Proc:eedings of theSubcontractors' Review Pileeiin,g

Aquatic Species Program

Sponsored and Organjzed by:

Solar Energy Research InstituteBiomass 'Program Office

for the

UaS. Department of Energy

July 1, 1981Washington, D.C.

NOTICE

This report was prepared as an account of the meeting sponsored by theUnited States Government. Neither the United States nor the UnitedStates Department of &,ergy, nor any of their employees, nor any oftheir contractors, subcontractors, or their employees, makes anywarranty, express or implied, or assumes any legal liability orresponsibility for the accuracy, completeness or usefulness of anyinformation, apparatus, product or process disclosed t or representsthat its use would not infringe privately owned rights.

This report has been reproduced directly from the best available copy.

r' ,

SERI/CP-624-1228

PROCEEDINGS OF THE

SUBCONTRACTORS' REVIEW MEETING

AQUATIC SPECIES PROGRAM

JULY 1, 1981

WASHINGTON. D. C.

SPONSORED AND ORGANIZED BY:

SOLAR ENERGY RESEARCH INSTITUTEBIOMASS PROGRAM OFFICE

FOR THE

u.S. DEPARTMENT OF ENERGY

TABLE OF CONTENTS

Year One Progress Report on Algal Production Raceway ProjectE. A. Laws, University of Ha~Jaii ••••••••••••.••••••••••••••• 1

Effects of Light Intensity and Nitrogen Deficiency on Yield,Proximate Composition, and Photosynthetic Efficiency ofPHAEODACTYLUM

w. H. Thomas, D. L. R. Seibert, M. Alden, and A. Neori,University of California, San Diego •.••••••••••••••••••••••• 33

Algal Research at SERl: A Basic Research of PhotobiologicalProduction of Fuels and Chemicals by Microalgae

S. Lien, Solar Energy Research Institute •••••••••••••••••••• 59

Paper Not Available at Time of PrintingDr. John Benemann,Ecoenergetics, Inc. •••.•••••••••..••••••••••••.•••••••••.••• 67

Assessment of Blue-Green Algae in Substantially Reducing NitrogenFertilizer Requirements for Biomass Fuel Crops

D. B. Anderson, P. M. Molton and B. Metting,Pacific Northwest Laboratory, Richland, WA. andR&A Plant/Soil, Inc., Pasco, WA. ••.••.••.••••••••••••••••••• 69

Water Hyacinth Wastewater Treatment SystemC. A. Lee, WED Enterprises, Glendale, CA andT. McKim, Reedy Creek Utility Company, Lake Buena Vista, FL.. 75

Cultivation of Macroscopic Marine AlgaeJ. H. Ryther, Woods Hole Oceanographic Institution,Woods Hole, 'MA.. ••••••••••••••••••••••••••••••••••••••••••••• 111

Wetland Biomass ProductionD. C. Pratt, University of Minnesota Botany Department,St. Paul, ~ • . ••••••.••••••••• ,. .........•.....••.••••.••. ".. 119

iii

YEAR ONE PROGRESS REPORTON ALGAL PRODUCTION

RACEWAY PROJECT

E.A. LawsUniversity of F..awaii

Oceanography Dept.2525 Correa Rd.

Honolulu, HI 96822

1

INTRODUCTION

The principal objectives of the Algal Production Raceway (APR) project aretwofold. First, we want to confirm that lipid and protein production canbe maintained over a six-month period at a sufficiently high rate to makethe project economically and energetically attractive. Second, we want todevelop a harvesting mechanism which can remove cells from the APR at asatisfactory rate and with a minimum of energy expendature.

Much of the culture work during the first year of the project has involvedsmall volume chemostats, since construction of the raceway systems was notcompleted until January, 1981. The chemostat work has allowed us tothoroughly a~mine the growth and compositional characteristics of Phaeodacty­lum tricornutum as a function of light and nutrient limitation, over a rangeof temperatures and as a function of light quality. These studies clearlyshowed that maximum lipid and protein production would be achieved underlight-limited conditions, and near the end of the first year a thorough studyof protein and lipid production under light-limited conditions in two strainsof P. tricornutum revealed a 40% and 70% difference in lipid and proteinproduction respectively between the ~~o strains. This difference in productionis quite important to the overall economics of the project.

During the first few months of operation of the APR, we were able to exploreseveral different cell harvesting mechanisms. At the moment the most promisingmethod appears to be collection of foam, which contains 10-40 times the celldensity in the APR and is produced in more-than-adequate amounts by aerationof the raceway when culture densities reach a few million cells per ml. Thefoaming is apparently the result of the release of surfactants by the cells.A compositional analysis of cells collected from the APR has revealed resultsvery similar to those reported earlier for P. tricornutum in small-scale APRprototypes.

The chief disappointment in the initial work on the APR has been our failureto produce cell densities above about 5x10 6 cells'ml-1 at growth rates compara­ble to those needed to make the APR concept economically attractive. Earlierresults from the APR had indicated that growth rates of about 0.4 d-1 could beachieved at cell densities of about 2.5x10 7 cells·ml-1 , but cell densities inour APR have never exceeded about 1.1x107 cells'ml-1 , and growth rates at thesehigh cell densities have been well below 0.4 d-1

• Even with cell densities of2.5x10 7 cells·ml-1 , our calculations indicates that an .~R would have to beabout 0.6 m deep in order to achieve maximum production. The system wasinitially envisioned as being operated with a depth of only a few em and at avery high cell density, but it appears that release of inhibatory metabolitesby the cells will prevent us from operating the system in that mode. We havetherefore redesigned the APR to accommodate a greater culture depth.

CAP~ON PRODUCTION BY P. TRICO~~Tu}l BASED ON CONTINUOUS CULT11tE RESULTS

Consider an APR system having a depth D meters in which P. tricornutum is grow­ing at a rate U ner day. If the concentration of particulate carbon in theraceway is C g'm~3, then the rate of production of carbon is uCD g·m-2d-1

• If

2

C' is the concentration of ChI a in the raceway, then attenuation of light withdepth in the culture can be well described by the equation

I = I e-KC'Z (1)Z 0

where 1Z is the light intensity at depth Z, 10 1s the surface light intensity,and K is the extinction coefficient of light per unit Chl a. The average lightintensity I in the water column is then

1 D 10 -KC'DI = D ! IZdZ =KC'D (1 - e ) (2)

oFor reasons which will soon become clear, we can assume the APR will be rununder conditions in which e-Ke'n «1, so that

I s 10KC'D

Hence production in the raceway can be written as

(3)

(4)CD = llC • c'n = 'f,lCIou C' C'KI

Most values reported for K fall in the range 10 to 20 m2 • g-l ChI a. In theactual APR however, we can a~pect that substances other than ChI a will scatterand/or absorb light. Experiments conducted by us indicate a K value of about23 at high cell densities, and measurements made in the APR in early May, 1981,yielded a K of 22. In modeling production, we have assumed a K value of22 m2 • g-lChl~.

Average incident solar radiation in Hawaii is estimated to be about 3.1xl01 0

BTUeacre-1·yr-1• Taking 50% of this figure to be visible light (400 - 700 nm),

assuming a 12h photoperiod, and using appropriate conversion factors, wecalculate the mean incident light intensity to be about 1200 llEin e m-2 • 8-1•

Mean light intensities recorded by us at the APR site during May, 1981, were·1150 lJEin em-2es-1 • We have assumed an 10 of 1200 in the following calculations.

1. Light-Limited Carbon Production

Figures 1 and 2 show plots of growth rate and the carbon:Chl!. ratio (C:C')as a function of I for the Woods Hole (WH) strain of P. tricornutum. Analogousdata have been obtained for the West Coast (Thomas) strain. Using the smoothcurves fit to these data, the plots of total carbon production as a function ofI were generated in Figs. 3 and 4. For both strains, carbon production peakedat ~~ I of about 35 ~Einem-2·s-1, but production was about 63% higher in theThomas strain. The principal difference in the two strains was in the C:C'ratio, which was much higher in the Thomas strain. Assuming that carbon accountsfor about 55% of the ash free drJ weight (AFDW) of P. tricornutum, theseproduction estima~es translate into 35.9 and 58.3 short tons F~DW per acre-year. The latter figure compares favorably with the estimate of 62.5 short tonsAFDW per acre-year in the original APR proposal.

3

2. Light-Limited Lipid Production

Figure 5 shows the percentage of cellular carbon as lipid as a function of Ifor the WH strain. An analogous plot was obtained for the Thomas strain.Lipid carbon production can be calculated as a function of I by multiplyingthe right-hand side of eq. 4 by the fraction of cellular carbon in lipid. Theresults for the WH and Thomas strains are shown in Fig. 6 and 7. Peak lipid Cproduction for the lVH and Thomas strains is estimated to be 15.8 and 22.5 tonsC 'ha-1 • yr-1 at a light intensity of 35 ~Ein • m-2 • s-l. Assuming thatcarbon accounts for about 74% of the AFDW of the lipid, these figures translateinto 9.5 and 13.5 short tons lipid AFDW per acre-year respectively. Assumingthat this lipid consists of 70% free oils with an energy content of 17,000 BTUper pound, and that a barrel of oil is equivalent to 6x106 BTU, the free oilproduction rate is calculated to be 37.7 and 53.7 bbl per acre-year respectively.

3. Light-Limited Protein Production

Figure 8 shows the percentage of cellular carbon as protein as a function of Ifor the WH strain. An analogous plot was obtained for the Thomas strain.Multiplying the right-hand side of eq. 4 by the fraction of cellular carbon asprotein yields the graphs of protein production shown in Figs. 9 and 10.Maximum production again occurred at a light level of 35 ~Ein • m-2 • s-l, andequaled 18.6 and 31.7 tons C • ha-1 • yr-1 for the WH and Thomas strainsrespectively. Assuming carbon to account for 51.3% of protein AFDW, thesenumbers translate into 16.2 and 27.6 short tons protein AFDW per acre-year.

4. Effect of Light on Production at a Fixed Nutrient-Limited Growth Rate

We consider here onlr the growth rate 0.41 d-1 • Although experiments were alsoperformed at 0.75 d- , the results at this second growth rate are qualitativelysimilar to the results at 0.41 d-1 • The calculations were made using eq. 4for total carbon production, and by multiplying eq. 4 by the fraction ofcellular carbon as lipid or protein to determine lipid and protein productionrespectively. For the sake of brevity, we include only the final results here.These results are summarized in Figs. 11-13.

It is clear from these figures that reducing the light level in a nutrient­limited culture (by either increasing the culture density or increasing thedepth of the culture), so as to ultimately produce a light-limited culture atthe same growth rate increases the rate of production per unit area of theculture. If this conclusion is true at all growth rate, then maximumproduction is obviously obtained in a light-limited system. In order to checkthis hypothesis, we performed a series of studies under nutrient-limitedconditions.

5. Effect of Nutrient Limitation on Production at I = 300 uEin -2• m

The results reported here were obtained in a manner similar to those previously

8

reported. }~ximum growth rate at I = 300 ~Ein'm-2's-1 is about 1.05 d-1 andin theory production curves based on nutrient-limited and light-limited datashould merge at that growth rate. Figures 14-16 show the relevant results forthe tvH strain. In all three cases, maximum production was achieved under lightlimitation, the optimal growth rate being about 0.4 d-1• The crossing of thetotal C and protein C production curves at growth rates slightly less than1.0 d-1 reflects errors in the curve fitting process; as noted the curves shouldmerge at a growth rate of 1.05 d-1•

6. Conclusion

The data not only indicate that protein and lipid production are maximized ina light-limited system, but that production of both is maximized at essentiallythe same light intensity. While it is true that the compositional characteristicsof P. tricomutum vary widely as a function of growth conditions, our resultsindicate that it would not be econimically or energetically reasonable to varythe growth conditions of P. tricomutum so as to change its composition, becausedeviation from optimal growth conditions reduces both protein and lipidproduction. Optimal production is achieved using the Thomas strain at a light­limited growth rate of 0.39 d-1, under which conditions about 31% of the cell'scarbon is lipid carbon and about 44% is protein carbon. From an examinationof the graphs in Figures 4, 7 and 10, it is apparent that a light intensityslightly below the optimal value can greatly reduce production, and given day­to-day variability in solar insolation it would presumably be desirable tooperate an APR somewhat to the right of the optimum light intensity point inorder to avoid large reductions in production on cloudy days.

EFFECT OF A CuSO~ FILTER ON LIPID AND PROTEIN PRODUCTION

Early work with the APR utilized a roughly 0.9 cm thick 3% CUS04 solution toabsorb infrared light and thus retard overheating of the culture. Someinformation suggests that lipid production is stimulated when cells are grownin blue light, and it therefore seems possible that use of CUS04 filters wouldserve a dual purpose in the raceway system. The experiments reported here wereconducted with the tVH strain, using a 10% CuSO~'5H20 solution about 0.25 cmthick contained in a plastic sandwich.

The results of the study, conducted under nutrient saturated conditions. areshown in Fig. 17. C:Chl a ratios and the percentage of protein carbon areincreased under blue light, and the percentage of lipid C is slightly reduced.Growth rate as a function of light intensity was virtually indentical underblue and white light. Since introduction of the CUS04 filter obViously removessome light quanta from the visible spectrum, an allowance must be made for thisfact in evaluating the effectiveness of the filter system in stimualting carbonproduction. Since our filter removed about 40% of incoming visible quanta.the conclusion from Fig. 17 is that at I = 35 ~Ein'm-2's-1 there would beessentially no net enhancement of lipid production. a 20% enhancement of proteinproduction and a 5% enhancement of total carbon production.

18

The conclusion is that while large scale enhancements of carbon productionare not likely to result from the use of CuSO~ filters, the filters canprobably be designed in a manner which will not reduce production, and willprobably lead to a modest increase. Since the filters reduce the managementcosts of the raceway be retarding overheating of the culture and storing apotentially valuable byproduct (heat), we conclude that installation of CuSO~

filters is advisable in the APR.

EFFECT OF FLASHING LIGHT ON CARBON PRODUCTION

The rationale for conducting these experiments is that efficiency of lightutilization is improved if light is provided in alternating light and darkperiods rather than continuously. The flashing light effect has been documentedfor time periods as short as 0.02 sec and as long as several hours. There isalso evidence that simply modulating the light intensity about a mean valuemay produce a 20-80% increase in production versus incubation at the meanlight intensity. Although the physiological mechanisms which underlie theflashing light effect probably differ over the wide range of time scales studied,the details of the mechanisms need not concern us. Simulation of a flashinglight or modulated light effect in the APR system could greatly enhanceproduction. The idea would be to grow a sufficiently dense and turbulent culture·so that vertical mixing would introduce the cells alternately into lighted anddarkened regions. Since we anticipated that the time scale of such mixingprocesses would be on the order of one second, our initial axperiments wereconducted on that time scale. The cells were exposed to light 50% of the time,and the results compared to those from cells grotvn on continuous light.

There was little difference in the composition of the cells grown under flashinglight and continuous light. The growth rate of the cells grown under flashinglight was substantially lower than that of the cells grown under continuouslight, but the difference was often less than a factor of two. Based on thecomposition and growth rate data, production curves under continuous light andthe flashing light regime are shown in Fig. 18. In making this comparison, wehave multiplied the flashing light production figures by a factor of two, on theassumption that during the dark period the light is used to illuminate a secondculture. In effect this calculation amounts to assuming that the culture isseparated into two subpopulations, which are alternately illuminated and darkenedas the culture is mL~ed. The results show that production is enhanced at lightlevels greater than about 125 - 150 ~Ein • m-2 • 5-1 under the flashing lightregime, but that there is a reduction in production at lower levels of flashinglight. The increase in production at 200 ~Ein • m-2 • S-l is only 10-15%.

While these experiments do not seem particularly encouraging from the stand­point of increasing production, it is possible that a different light:darkperiodicity could significantly change the picture. The present results forexample indicate that for I = 1000 uEin • m- 2 • s-l, lipid production under aflashing light regime would be about 70% greater than lipid production undercontinuous light t and it is therefore conceivable that exposing populations toshort pulses of very bright light followed by relatively long periods ofdarkness could substantially increase production.

23

EFFECTS OF TEMPERATURE ON CARBON PRODUCTION

Growth rate and composition data collected from nutrient-limited populationsof P. tricomutum indicate that a temperature of 26.5°C is nearly lethal toP. tricornutum, while a temperature of 24.S oC appears close to optimal in termsof maximizing growth rate. Overheating can obviously be a serious problems ina shallow, dense culture. On the other hand, an effect to cool the raceway toa temperature as low as 20°C for ~~ample would be costly. It is thereforedesirable to know how protein and lipid production are affected by temperature.

Our results show that the C:Chl a ratio in the cells is negatively correlatedwith temperature, but that there-is very little correlation between eitherpercent lipid or percent protein and temperature in the range 21.5°C ~ T ~

26.5°C. As a result, carbon, protein and lipid production are all negativelycorrelated with temperature, as indicated for lipid production in Fig. 19.These experiments were carried out under nutrient-limited conditions, and itis not clear what effect temperature will have at low light levels (~35 ~Ein •m-2 • S-l) and nutrient-saturating conditions. However, the data in Fig. 19indicate that at high light levels (~300 ~Ein • m-2

• s-l) and nutrientsaturating conditions (~~1.2 d-1

) there is about a 36% increase in lipidproduction if the temperature is reduced from 24.5°C to 21.5°C.

The optimum temperature at which to operate APR's will eVidently represent atrade off becween cost of temperature control and the value of increasedproduction. Although temperature control beyond the use of !R filters may beuneconomical in production APR's, temperature considerations may influence thegeographical siting of APR's. Furthermore, if a cheap cooling mechanism isavailable (e.g. DTEC water effluent from a land-based OTEe facility), thenmore active temperative regulation may be advisable.

APR RESULTS

Table 1 summarizes a few of the results obtained with the APR during a five­day period from 3/18/81 to 3/23/81. During this time period, the APR wasdiluted at a rate of 0.2 d- 1

• From the changes in cell numbers, particulatecarbon, particulate nitrogen and dry~ight, we conclude that the cells weregrowing at a mean rate of 0.32 - 0.38 d- 1 during this period. These figurescompare reasonably well with the mean growth rate of 0.39 d- 1 reported fromearlier APR prototypes. Using the % protein versus N/C regression in Fig. 20and the mean N/C ratio of 0.214 from the APR product in Table 2, we concludethat the APR cells' carbon consisted of about 55% protein carbon. Assumingthat protein C accounts for 51.3% of the AFDW of protein and that C accountsfor 54.93% of the AFDW of P. tricornutum, we conclude that about 58% of theAFDW of these cells consisted of protein. This figure compares well with thevalue of 63% reported from earlier APR prototypes. An analysis of percent ashon APR samples gave a figure of 37% as compared with a mean value of 35% fromthe earlier prototype.

Thus in terms of growth rate and cellular composition, the cells in our APRappear very similar to those in earlier APR prototypes. The principaldifference in the ~NO cultures has been in the cell densities. Cell densities

in our APR have generally been abou~ Sxl0 6 cells-ml-1, while earlier APR's

were operated at cell densities of 20-30xl0 6 cells-ml-1•

Let us assume for the moment that we are able to achieve a cell density ofabout 25xl0 6 cells per ml in the APR. How deep a system would be required toachieve the optimal light intensity of about 40 ~E·m-2·s-1? Based on the datain Table 1, at a cell density of 25xl0 6 cells·ml-1 we can a~pect a ChI aconcentration of about 2250 mg-m-3 • Given this ChI a concentration and­assuming a specific extinction coefficient of 22m2.g~ ChI a. the extinctioncoefficient of light in the racway should be about sOm-1 at-a cell density of25xl0 6 cells·ml-1 • If the incident light intensity is 1200 ~E·m-2·s-1,reduction of the water column mean light intensity to about 40 ~E·m-2·s-1 wouldrequire a water column 1200/(40·50) = 0.6 m deep. Assuming a surface area forour APR of 50m2 , we have a total volume in the APR of 3000 liters.

If we assume the population is growing at about 0.4 d-1 , the optimum growthrate based on preVious calculations, then we need to harvest the cells from(0.40)(3000) = 1200 liters of water per day. We have found that foam isconstantly produced in the APR as a result of the aeration system which drivesthe circulation_ The density of cells in this foam is about 40 times thedensity of cells in the APR. Therefore we need to harvest 1200/(40-24) =1.25 liters-hr-1 of foam. We have found thata simple commercial liquid vacuumsystem will remove over 2~ per hour of foam if operated about 10% of the time.Thus the amount of foam generated at a single point in the raceway appears morethan adequate to provide the harvesting mechanism we need. The materialcollected from this foam can be easily dried in the sun in a matter of hoursto yield a product suitable for harvesting.

We therefore feel that we have identified a suitable harvesting mechanism.There is no doubt that generation of foam can be effected at numerous pointsin the APR by means of air bubbling, and the efficiency of foam production canbe increased with the use of air stones to produce fine bubbles_ At presentthe foam is trapped with a weir device and removed by periodic suction, butit is likely that in a production APR a skimmer would prove to be more energyefficient in harvesting the foam than our suction device. We have continuedto use the suction system for the present, as it is more easily adapted tochanges in APR design.

ECONOMIC ANALYSIS

Our chemostat results indicate that under optimum production conditions wecan expect to produce 0.92 bbl of free oils and 0.47 short tons of protein forevery short ton of AFDW produced by the APR. Costs of operating mass culturesystems vary greatly, from about $800 per acre-year to $18,000 per acre-year.The original APR proposal assumed an operating cost of $12,401 per acre-year.Table 2 shows the projected break-even prices of the oil product assumingoperating costs of $12,401 and $18,000 per acre-year as a function of totalbiomass production (AFDW). In making this analysis, we have assumed a valuefor the protein byproduct of $248 per short ton as in the original APR proposal.

The results indicate that at an operating cost of $12,401 per acre-year, oil

29

TABLE 2. BREAK-EVEN COST OF OIL PRODUCED IN APR SYSTEM. COLUMNS A AND B ASSU~.E OPERATING COSTS OF$12,401 and $18,000 PER ACRE-YEAR RESPECTIVELY.

Total Production Protein Production Oil Production Break Even(tons AFDW/acre-year) (tons AFDW/acre-year) (bbl/acre-year) Cost($/bbl)

A B

20 9.4 18.4 547 852

30 14.1 27.6 323 526

40 18.8 36.8 210 362

w 50 23.5 46.0 143 2650

60 28.2 55.2 98 199

70 32.9 64.4 66 153

80 37.6 73.6 42 118

90 42.3 82.8 23 91

100 47.0 92.0 8 69

can be produced at a cost comparable to today's prices if total production isapproximately 80 short tons AFDW per acre-year. If production costs reach$18,000 per acre-year, then competitive prices would not be achieved even ata total production rate of 100 short tons AFDW per acre-year. At this pointone may be guardedly optimistic in assuming that a total production rate of 80short tons AFDW per acre-year can be achieved. This figure translates into 179metric tons AFDW per ha-yr, 2-3 times the levels of production typicallyachieved in mass culture systems. However, the figures is only about 35% higherthan the optimum production based on chemostat work. It is possible that theeffects of modulated blue light on the system will allow the extra productionto be achieved.

31

EFFECTS OF LIGHT INTENSITY AND NITROGEN DEFICIENCY ON YIELD,PROXIMATE COMPOSITION, AJ.'ID PHOTOSYNTHETIC EFFICIENCY OF PHAEODACTYLUM

w. H. Thomas, D. L. R. Seibert, M. Alden, and A. NeoriInstitute of Marine Resources

Scripps Institution of OceanographyUniversity of California, San Diego

La Jolla, CA 92093

ABSTRACT

Phaeodaa~dZum tricornutum Bohlin was grown in a large 5-cm thick culturecontainer in seawater enriched with N, P and trace metals. The culture wasdense (~lOOO mg dry weight liter-i) and was grown at three light intensities(39.5, 59.6, and 69.8% of maximum sunlight at La Jolla) in the batch culturerode. The lowest intensity gave the highest yield (21.74 gut m-Zday-1) andefficiency of photosynthetically active (400-700 rum) light utilization(12.23%). Other intensities were slightly or greatly photoinhibitory. Con­tinuous culture under N-sufficient conditions resulted in a somewhat loweryield and efficiency than in batch culture. The proximate composition ofN-sufficient cells was about 60% protein, 10% carbohydrate, 20% lipid and10-17% ash. Nitrogen deficiency greatly reduced yield, efficiency, and pro­tein content. Lipid content was increased to 30% of the dry weight. Weconclude that, because of the reduced overall cellular yield, increasedlipid content does not result in an increased lipid yield in Phaeodaaty~um.

These results are compared with other investigations both in the laboratoryand in outdoor mass cultures.

INTRODUCTION

One of the general goals of the Solar Energy Research Institute's AquaticBiomass Program is to promote the use of microalgae as a source of energyand organic products. Outdoor mass cultures are envisioned in which algaewill utilize sunlight at a high efficiency as compared with land plants (seeWassinck et aZ.~ [1]). Much research on the production of valuable cellularproducts by mass culture has been performed with freshwater microalgae,principally Ch~oreZla [2, 3, 4, 5], but little work has been done on ~~e

capabilities of marine microalgae. This is somewhat surprising since sea­water is an abundant source of water and certain major nutrients (K, Mg, Ca,S, etc.) and coastal areas, partiCUlarly in the tropics, often receive agreat deal of sunlight on a yearly basis. Furthermore, marine microalgaemay be adaptable to growth in other sources of saline water such as thoseoften found in desert areas.

The purpose of our project is to study marine rnicroalgae (phytoplankton) inthe laboratory to provide essential basic info~ation useful for establish­ing successful outdoor mass cultures that ~Nill produce a maximum amount ofcellular energy and valuable byproducts. We are mimicing, insofar as pos­sible, natural sunlight, and are screening a number of marine species for

33

their ability to incorporate light energy at high efficiencies into a cel­lular product that is high in total energy (calories) and in lipid. We aretesting the effects of nitrogen deficiency and, to a certain extent, lightintensity, in manipulating the cells toward these goals. In addition toanalysis of harvested cells for total lipid and caloric content, we areanalyzing for other proximate constituents such as protein, carbohydrate,carbon, nitrogen, and ash.

This paper reports results with the first marine microalgal species that wehave investigated, Phaeodaatylum triaornutum Bohlin. We chose this speciesbecause it is easy to grow relative to others; a great deal is already knownabout its biology, chemical composition, physiology, and biochemistry; andit is being used as a test organism by another SERI subcontractor in Hawaii.

~1ATERIAL AND METHODS

The Microalga

We have cultured Phaeodactqlum tx-ieomutum in our laboratory for many years.This alga was provided to us by Dr. David Leighton and probably came fromthe University of California Marine Laboratorj at Bodega Bay. However, itsprecise origin and strain designation are ~~nown.

Medium

Phaeodaa~ylum was grown in filtered (0.45-~ pore size) Scripps Pier sea­water enriched with 15 mg-at N liter- 1 as (NH4)ZS04' 1.25 mg-at P liter- 1 asKzHP04, and the trace metal mixture in medium "f" [6). For N-deficientmedi~~ the (NH4)zS04 enrichment was deleted.

Preculture

Prior to the main experiment, the alga was precultured to a high density ina nine-liter serum bottle incubated at 21°C and a continuous, incident lightintensity of 0.056 cal cm-Zmin- 1 in the photosynthetically active spectralrange (400-700 nm). The light was supplied by a bank of fluorescent lamps.The culture was bubbled with a 1% CO -in-air at a rate of 2000 mt min- 1

6 Z -1After a cell density of 6.72 X 10 cells rot was reached, 3.4 liters ofthis preculture was poured aseptically into the main culture apparatus.

Light Supply

The main culture apparatus is diagrammed in Figure 1. The light source wasa 2000-watt tungsten-halide lamp (3200 0K) mounted in a 50-em diameter para­bolic reflector such as is generally used for stage lighting. The la~p wasope~ated diurnally (12:12 LID cycle). wben the lamp was turned on, voltagewas applied gradually over a one-minute period and was controlled by an auto­matic dimmer connected to a tirr~ng circuit. This prolonged lamp life beyond~~e rated 500 hours to >1100 hours so that one bulb was used throughout the

34

experiment (94 days). The light was filtered through a 7-em thickness offlowing tapwater and a 3-cm thickness of 3% Cuso~ to remove infra-redradiation. Figure 2 shows spectral distributions (350-800 nm) of lightenergy (3 intensities) at the culture surface with and without the CuS04filter. The latter filter removed red and far-red irradiance quite effec­tively. These spectral distributions were measured with an Optronics 74l-Vspectroradiometer. Light intensity was not uniform over the whole surfaceof the culture1 at the medium light level it varied from 0.095 to 0.247 calcm- 2min- 1 in the photosynthetically active spectral range (400-700 nm) asmeasured at 143 points over the culture area with aLi-Cor 190 SE flatresponse energy sensor. These 143 intensities were integrated over thewhole culture surface area (705.3 cm2

) to obtain values for total caloriesreceived by the culture per 12-hour light period. These values were 62,715,94,661, and 109,405 calories per culture per light period for the threelight intensities used. In 1967 Strickland [7] measured total daily sun­light radiation at La Jolla from April to September. The maximum value was450 cal cm- 2day-l. Over the area of our culture (705.3 cm2 ) this would be317,389 cal day-l of which one-half (158,695 cal day-I) would be in the 400­700 nm spectral range. Dividing the cal day-l figures for low, medium, andhigh intensities for the culture by 158,695 cal day-l yields values for thepercentage of maximum sunlight at La Jolla supplied to our culture. Thesewere 39.5, 59.6, and 68.9%, respectively, for the three intensities. Themedium light intensity was used for most of the experiment. The lower in­tensity was achieved by placing a plastic diffusing screen over the CuSO~

filter; and the higher intensity was achieved by moving the light source to8 em from the tapwater filter rather than the 20 em distance shown inFigure 1. The spectral distribution of the light was not changed by varyingthese intensities (see Figure 2). At a given location at the culture sur­face light intensities integrated from 400-700 nmby the spectroradiometerwere within 5% of those measured with the Li-Cor sensor. Daily measurementsat the location of maximum intensity (I o) with the Li-Cor sensor establishedthat the lamp output did not vary appreciably «10%) over the 94-day experi­ment. Measurements (Li-Cor sensor) were also made of light (I) transmittedthrough the 5 cm culture thickness to roughly follow culture growth. Duringthe experiment the logarithm of IlIa correlated very well with cellular dryweight per liter (r = 0.856).

The Culture

Figure 1 shows the side configuration of the main culture container whichwas constructed of clear plexiglass. It was 5 em thick, 39 em deep, and24 cm wide. The bottom was curved for ease of mixing. The container wasfilled to a depth of 28 cm with 3.4 liters of preculture, after steriliza­tion of the container with 70% ethanol followed by a rinse with sterile de­ionized water. An aeration (1% CO2-in-air at 2000 mi min- 1 ) tube reachedthe bottom of the culture and aeration mixed the culture vigorously. Thegas was washed by bubbling through sterile deionized water. An overflow andpressure-release tube extended to the culture surface; overflow regulatedthe culture volume at 3.4 liters. A sampling tube extended into t~e centerof the culture. Sampling and harvesting we=e carried out by clamping off

35

FIGURE 1. DIAGRAM OF CONTINUOUS CULTURE APPARATUS.

36

FIGURE 2. SPECTRAL DISTRIBUTION OF FILTERED 3200 0K LIGHT SUPPLIED TO !<fAINCULTURE. A: LOW INTENSITY, 62,715 CAL DAy- 1 ; B: MEDIUM INTENSITY,94,661 c.1U. DAy-l; C: HIGH INTENSITY, 109,405 CAL DAy-l. THE DASHED LINESHOWS THE DISTRIBUTION OF B WITH THE CUSOl+ FILTER REMOVED AND WITH FILTRA­TION ONLY THROUGH FLOWING TAPWATER.

38

the overflow tube so that air pressure forced the sample out into a collec­tion container. A fourth tube did not reach into the culture and was for theaddition of medium via peristaltic pumping from a 20-liter glass carboy. Themedium consisted of sterile-filtered Scripps Pier seawater enriched as de­scribed above. It was continuously mixed by aeration. The culture andmedium were in contact only with plexiglass plastic, Pyrex glass, and sili­cone rubber. While temperature-controlled water was circulated through acooling jacket surrounding the sides of the culture container, this was notadequate to control culture temperature and temperature was principally con­trolled by placing the entire apparatus, including the lamp, in a cold roomset at 20°C. During the 94-day experiment the culture temperature was 22.7± 1. QOC as measured daily two hours after the light came on. Because tempera­ture control could only be achieved with the cold room air-conditioner, wedid not attempt to vary temperature diurnally.

Sampling, Harvesting, and Growth Measurements

Daily sampling for routine growth measurements was carried out over the full94 days of the experiment by withdrawing <100 mt of cell suspension. peri­odically larger amounts (300-700 rot) were withdrawn as harvests for proximateand caloric analyses. Portions of each harvest (240 rot or greater) werecentrifuged at 1270 g for 20 minutes. The centrifuged cells were washed oncewith deionized water and dried at 60°C. Other portions of the harvest orsample were filtered on glass fiber filters for proximate analyses and dryweight determinations. The dry weight samples 'Here dried at 60°C and theother samples were frozen for later analysis.

Growth was measured by cell numbers, optical density, and dry weight. Cellswere counted in diluted (filtered seawater) suspensions with a Model BCoulter Counter. Optical density was measured at 750 nm in a 1 em spectro­photometer cell. For dry weights, filters were weighed on a Cahn micro­balance. The weights were corrected for adsorbed salts by weighing blankfilters through which filtered seawater had been run. Cell numbers corre­lated well (r = 0.821) with dry weights throughout the experiment as didoptical densities (r = 0.867). pH of the cell suspension was measured dailyor oftener. Due to NHl++ uptake, pH was sometimes reduced to values <5.0.It was adjusted back to levels >7.0 by addition of 1M KOH.

Analvtical Met~ods•

Total cellular carbon, nitrogen and hydrogen analyses were performed with aHewlett-Packard Model 185 CHN analyzer [8]. Cellular protein was calculatedfrom the %N value X 6.25. Cellular carbohydrate was analyzed by the phenol­H

2 SOl+ spectrophotometric method of Myklestad and Haug [91. To ensure thatpolysaccharide was hydrolyzed completely by the 80% H2S04 , the samples wereheated for one hour at 10QoC and then allowed to stand at room temperaturefor 24 hours. Lipid was extracted from frozen filtered cells by 2:1 (V/V)methanol-chloroform at 40°C [101. The extract was washed with deionizedwater to remove water-soluble material and an aliquot was dried at 40°C in ateflon cup and weighed on the microbalance. Considerable methanol-chloroform

40

extractable material was removed by the water wash and the lipid values wouldhave been much higher if the wash had not been done. After their initialweighing, dry weight filters were heated at 450°C overnight and reweighed todetermine ash content. Ash was also determined by weighing the residue fromthe combustion of caloric samples and by combustion of 5-10 mq pellets ofcells at 450°C. Caloric values were determined with a Phillipson microbombcalorimeter in dried, centrifuged cells. The calorimeter was connected to arecorder and to a millivoltmeter and the apparatus was standardized by com­busting known weights of benzoic acid. Cells were filtered and frozen forpigment analyses, but these have not yet been completed. During the develop­ment of N-deficiency (attained by pumping in N-free medium), NH

4+ in the

supernatant from centrifuged samples was measured with a Lazar NH 4 + electrode.

Analytical Precision

Five separate samples were filtered for each analysis. Precision (standarddeviation/mean) was 6.3%, 5.8%, 18.9%, 18.6%, 11.9%, <1%, and 2.3% foranalyses of carbon, nitrogen (and protein), hydrogen, carbohydrate, lipid,cell numbers, and dry weight, respectively. Only a few replicates weredetermined for calories and ash. The "±" values reported are standard devia­tions.

RESULTS

Light Intensity Effects on Batch Culture eN-Sufficiency)

Initially the culture was grown in batch mode without dilution at a lightintensity chosen to approximate 50% sunlight. The actual intensity wasabout 60% of the maximum smnmer intensity at La Jolla [7]. At the end of theexperiment the effects of two additional intensities were tested.

Figure 3 shows increases in dry weight at the three intensities. Increasesin cell numbers and optical density were also linear. Table 1 gives thevarious culture parameters at these intensities. The maximum growth rate andyield occurred at the lowest intensity; these decreased slightly at themedium intensity; and apparent photoinhibition occurred at the highest inten­sity. The efficiency of light utilization was maximal at the lowest inten­sity and decreased greatly with increasing intensity. Although yield de­creased only slightly at the medium intensity, the efficiency was much lessthan that at low intensity since more light ~Nas probably being supplied thancould be used and there may have been slight photoinhibition at this mediumintensity.

At the three light intensities there were no large differences in the proxi­mate chemical composition of the crop nor in caloric content. The yield oflipid at low, medium, and high intensities was calculated to be 5.63, 4.11,and 3.21 gm m-2day-l, respectively; the yield of protein was 12.98, 12.08,and 10.05 gm m-2day-l, respectively, at the three intensities.

41

FIGURE 3. GROtV'l'H OF PHAEODACTYLUM AT THREE LIGHT INTENSITIES. A: AT62,715 CAL. DAy-l; B: AT 94,661 CAL. DAy-I; C: AT 109,405 CAL. DAy-l.THE GROWTH CURVES ARE DISPLM:ED FOR CLARITY.

42

TABLE 1. PH1>EODACTYLUM BATCH CULTURE PARAMETERSAT THREE LIGHT INTENSITIES AND NITROGEN SLTFICIENCY.

ABatch Culture

B cCulture Dates

Light**:

2/28-3/2/81 12/12-12/14/80 3/3-3/5/81

IntensityCal culture-1day-l% of maximum La Jolla sunlight

Growth:

Rate (mg liter-1day-l)Yield (grn m-2day-l)

Proximate Composition:

Carbon (% dry weight)Nitrogen (% dry weight)Hydrogen (% dry weight)Protein (% dry weight)Carbohydrate (% dry weight)Lipid (% dry weight)Ash (% dry weight)

Energy Content:

Calories rng d~ weight- 1

Calories mg C-

Efficiency**:

Low62,71539.5

45121. 74

48.4 ± 5.99.6 ± 0.814.0 ± 3.459.7± 5.313.0±3.125.9 ± 3.116.9

4.9710.27

12.23

Medium94,66159.6

43921.16

46.8 ± 3.5*9.1±0.6*15.3 ± 6.9*57.1::3.6*10.5±1.4*19.4±2.1*10.6*

5.09*10.55*

8.09

High109,40568.9

35417.09

42.6 ± 1. 79.4±0.823.9 ± 19.958.8±5.311.3 ± 1. 718.8::0.915.5

4.9311.57

3.49

* Proximate composition and calories from continuous culture(N sufficiency) on 12/29/80 and 1/5/81 samples.

** Light values are for photosynthetically active radiation(400-700 nm) and efficiencies are calculated only for thatpart of the spectrum.

44

Continuous Culture (Medium Light Intensity, N-Sufficiencv, 0.48 Volume Dav- 1)

Following the batch culture at medium light intensity, N-enriched medium waspumped into the culture at a nominal rate of 0.5 culture volume day-l and theculture overflow was collected. This part of the experiment lasted 19 days.The mean overflow volume was 1628 ± 99 mi day-l; this gave an actual dilutionrate of 0.48 volume day-1. The mean dry weight of cells was 702 ± 47 mgliter-I. The mean 0itical density was 2.23 ± 0.42 and the mean cell numberwas 12.47 ± 1.61 X 10 cells ~-l. From the overflow volume and dry weightvalues a yield of 16.20 gm m-2day- l was calculated. The proximate composi­tion of the cells was 48.2 ± 2.6%C, 9.4 ± O..2%N, 17.9 ± 5.5%H, 15.5% ash, 58.9±l.l% protein, 11.2 ± 0.8% carbohydrate, and 19.9 ± 3 ..5% lipid. The yield ofprotein, carbohydrate, and lipid would thus be 9.54, 1.81, and 3.22 gm m-2

day-1, respectively. The caloric value of the cell material was 5.09 caltug-1dry weight- I • The culture produced 1143 mg dry weight day-1 or 5813cellular calories day-1 and adsorbed 99 .. 0% of the light that was supplied or93714 calories day-l.. Thus the photosYnthetic efficiency was 6.20%.

At medium light intensity with N-sufficiency, batch culture gave a higheryield and efficiency than continuous culture -- 21.21 gm m-2day-l and 8.09%.We might have been able to dilute the culture slightly faster and achievestill higher yields and efficiencies in the continuous mode of operation, butdetermining the proper balance between dilution with N-enriched medium andgrowth to maintain a maximum yield is difficult when growth proceeds in alinear manner. The dilution rate for matching continuous culture to batchculture is highly dependent on the dry weight liter- I At the crop weightwe found (702 mq liter-I) the culture could have been diluted at 0.625 vol­umes day-1 to attain a yield equivalent to that in batch culture. Later onin the experiment, diluting at -0.75 volume day-l resulted in washo~t of thecells so this latter rate was too high.

Development of N-Deficiency (Medium Light Intensity, Continuous Culture, 0.48-1 )Volume Day

Nitrogen deficiency was induced by diluting the culture at the same rate asbefore with medium to which no NH

4+ was added. A~ Day 1 the NH~+ concentra­

tion in the culture medium was 11.0 mg-at N liter 1 (see Figure 4). Sincethe complete medium contained 15 mg-at N liter-l, the difference should havebeen 4 mg-at N liter- 1 in the cells. Calculation of the cellular N from thedry weight and %N value for this day gave a value of 4.1 mg-at cellular N1iter-1. Thus there was good agreement in total N in t.1-t.e cells as determinedin two independent ways. The NH u+ concentration in solution decreased ex­ponentially to zero «0.0001 mg-at NH~+-N liter-I) at Day 4 of deficien~J(Figure 4). .

Figure 4 also shows changes in crop parameters. Dry weight and cell nlZt'lbersremained nearly constant until the ~ri~+ was used up and thereafter decreased.Optical density decreased thrcughout the period of dilution r.-lith -N medium.

4S

FIGURE 4. CHANGES IN PHAEODACTYLUM CROP PARAMETERS AND IN NH4+ CONCENTRA-

TION DURING THE DEVELOPMENT OF NITROGEN DEFICIENCY (MEDIUM LIGHT INTENSITY,CONTINUOUS CULTURE AT 0.48 VOLtJMES DAy-l) •

46

Table 2 shows the yields as deficiency developed. These were determined bymultiplying the dry weight liter- 1 by the overflow volume in liters. Yieldsdecreased greatly after the NH~+ level in solution dropped to zero and cellu­lar N decreased. Also shown in Table 2 are cellular caloric values and effi­ciencies of light utilization. Caloric content increased somewhat when N­deficiency was pronounced due to increased lipid content (see Figure 6), al­though the caloric increase was not as dramatic as the decrease in yield andN content. Calories per unit carbon did not change with N-deficiency. Themean value was 9.93 ± 1.52 calories mg C- 1 during the whole development of de­ficiency. Efficiencies of light utilization were calculated from the yield,caloric content, and light absorbed by the culture. Efficiencies did notchange greatly with decreasing cellular N since caloric content increased asyield decreased. The mean yield of lipid in N-sufficient cells (Days 1-4)was 2.95 gm m-2day-1; that in N-deficient cells (Days 5-7) was 2.69 gm m-2

day-1 Thus even though the lipid content increased with N-deficiency(Figure 6), the yield of lipid was about the same as in N-sufficient cells.However mean protein yield decreased greatly; it was 7.59 gm m-2day-1 in N­sufficient cells and 2.34 gm m-2day-1 in deficient cells.

Figures 5 and 6 show the changes in cellular proximate composition with thedevelopment of N-deficiency. Cellular N and protein decreased greatly afterNH~+ was used up (Day 4). Cellular C did not change greatly nor did ash con­tent. Cellular carbohydrate increased significantly (p < 0.01) on Day 6 andthen decreased. Cellular lipid increased significantly (p < 0.01) on Days 6and 7. With N-deficiency, photosynthetically fixed carbon may be shuntedfirst into carbohydrate and then into lipid, in contrast to the formation ofthe main metabolic product, protein, in N-sufficient cells. The increasedlipid content and corresponding degree of reduction in the cells was paral­lelled by an increase in cellular H on Day 7. Extreme deficiency (Day 12)did not result in maintenance of a high lipid or H content even though cel­lular N was further decreased, and caloric content was also increased. Calo­ric samples from that day were "oily to the touch" and we cannot explain whythe lipid and H contents did not remain high.

In healthy cells, the cytoplasm was dispersed throughout the cell while in N­deficient cells the cytoplasm was clumped in the center of the cell. Oildroplets were possibly present in N-deficient cells, but we are not entirelysure of this point. Cell size did not change with N-deficiency nor did dryweight cell- 1 The latter was 56.6 ± 8.4 picograms cell- 1 in N-sufficientcells and under -N conditions it was 50.5 ± 9.2 picograms cell- 1 In all ofour microscopic observations cells THere single and fusiform; we never sawtriradiate cells.

Recovery From N-Deficiency

To allow the cells to recover from deficiency we pumped N-enriched mediuminto the culture. Recovery was slow and grow~~ did not keep up with dilutionfor several days. Therefore we stopped p~~ping and allowed the cells to re­cover under batch culture conditions. It still took 10 days for the cells tofully recover, at which time cellular composition was nearly identical tothat measured earlier in healthy cells.

48

TABLE 2. YIELDS, CALORIC CONTENT, AND EFFICIENCIES OF LIGHT UTILIZATION AS PUAEODACTYLUMCONTINUOUS CULTURES WERE SHIFTED TO NITROGEN DEFICIENCY i

(MEDIUM LIGHT INTENSITY, 0.48 VOLUMES DAy-I).

Date Day Cellular Calories Calories Efficiency Yield Lipid ProteinN Content mg dry mg C- 1 (\) (gm dry Yield Yield

(% of -1 -2(gIn dry (gm dryweight weight m

dry \"eight) -1 ) weight m- 2 weight m- 2dayday-I) day-1)

1/5/81 1 9.1 4 .. 92 10 .. 51 5.43 14.61 2.91 8.34

1/6/81 2 9.6 3.37 6.17 3.50 13.70 2.73 8.51

1/7/81 3 8.7 4.39 9.75 4.81 14.48 2.82 7.83

1/8/81 4 5.7 4.24 9.98 5.01 15.85 3.72 5.64

.f:- 1/9/81 5 3.2 3.69 11.46 4.05 13.96 2.79 2.83\0

1/10/81 6 3.9 5.53 10.93 4.91 10.43 2.89 2.51

1/11/81 7 3.5 5.48 10.15 4.09 7.75 2.39 1.67

1/16/81 12 2.5 5.73 10.48 * * * *

* Extreme N-deficiencYi no dilution (pumping) ; no growth; yields and efficiency could notbe calculated.

FIGURE 5. CHANGES IN %C, N, AND H DURING THE DEVELOPMENT OF NITROGEN DEFI­CIENCY IN PHAEODACTYLUM (MEDIUM LIGHT INTENSITY, CONTINUOUS CULTURE AT 0.48VOLUMES DAy- 1 UNTIL DAY 7 -- SEE ARROt'1).

50

FIG"URE 6. CF.ANGES IN % PROTEIN, CARBOHYDRATE, LIPID, AND ASH DURING THE DE-VELOPMENT OF NITROGEN DEFICIENCY IN PHAEODACTYLUM (MEDIUM LIGHT INTENSITY,CONTINUOUS CULTURE AT 0.48 VOLUMES DAy-l UNTIL DAY 7 -- SEE ARROW).

52

Continuous Culture at 0.25 Volumes Day-1 Under +N and -N Conditions (MediumLight Intensity)

Following recovery we pum~ed +N medium into the culture at an actual dilutionrate of 0.25 volumes day- to see if dry weight liter- 1 would increaseenough to increase yield over that at 0.48 volumes day-1. Dry weight did in­crease to 1045 ± 16 mg liter-lover that observed at 0.48 volumes day-1 (702 ±47 ~? liter- 1) but the yield at t~~ ne! dilution rate was on~1 11.59 gm m-zday as compared with 16.20 gm m day 1 at 0.48 volumes day .

We then took the culture through an N-deficiency cycle by diluting with -Nmedium at this new rate. Full deficiency took an additional day to achieve(to %N=3.l) as compared with t.~at during the former dilution rate and theeffects of deficiency were not as marked as before. That is, lipid increasedfrom 19.7% of the dry weight to only 23.2% and caloric values increased from4.91 calories mg-1 to 5.31 calories mg-1 Protein, of course, decreasedgreatly -- from 58.3% to 19.7%, but other constituents did not change appre­ciably. Yield went from 11.59 gm m-Zday-1 (+N) to 7.11 gm m-Zday-1 (full de­ficiency). Efficiency of light utilization went from 4.16% (+N) to 3.19%(-N) •

It seems apparent that diluting at this rate was very much sub-optimal ascompared with 0.48 volumes day-1 (+N conditions) where the yield was 16.20 gmm-Zday-1 and the efficiency was 6.20% (see above).

DISCUSSION

When we planned this experiment, we opted for a light intensity close to 50%of that of sunlight, since this was what might be received by outdoor masscultures screened by some sort of an infra-red absorbing filter, i.e. a CuSO~

solution. The actual intensity during most of the experiment was about 60%of the :naximum summer sunlight intensity at La Jolla. This was probablyslightly inhibitory and definite photoinhibition was found at an intensity of69% of sunlight (see Figure 3 and Table 1).

Mann and Myers [11] determined that photosynthesis (Oz evolution~ 1 em culturethickness, 40 mg drj wei~ht liter- 1)in FhaeodaatyZum was saturated at an in­tensity of 0.057 cal cm- min-lor for our culture vessel area and over a

-1 -1whole 12-hour day, 28,945 cal culture day Goldman [12J gives a generalcurve (his Figure 11, top curve) for yield versus light intensity at a simi­lar saturation intensity (Is = 0.06 cal cm-Zmin- 1

) , but in that curve does notconsider photoinhibition. At our low intensity (62,715 cal cUlture-1day-1)the yield- from his curve would be approximately 22 gm m-Zday-1. In batchculture we obtained a yield of 21.74 gm m-2day-1 at this intensity. It isprobable, that in our well-mixed, 5-cm thick culture with a drj weight approx­imately 25 times that used by Mann and Myers, the mean light seen by the cellswas very close to the saturation intensity they found for a thinner, lessdense culture. The agreement in these comparisons is remarkable since theymeasured short-term photosynthesis and we measured increases in dry weight

54

over a two-day period. Using Goldman's curve to calculate yield at ourmedium and high intensities, the values would be approximately 30 and 35 gmm-zday-1, respectively. We attribute our lower yields (Table 1) to photo­inhibition. Further empirical studies are needed on the effect of intensityon yield and efficiency in dense cell suspensions with a culture vessel suchas ours (or in a thicker culture vessel) i such studies would have obviousapplications to outdoor mass cUlturing.

Goldman [13] has summarized maximum and average yields of various algae inmany outdoor mass culture studies up to 1977. The maximum values range from

-2 -1 -2 -111-35 gm m day and the average values range from 2-27 gm m day • In a-2 -1later paper [14] he states that maximum yields of 30-40 gIn m day can be

expected under ideal conditions. A maximum yield of 3S. 6 gm m- 2 day-1 forsewage-grown algae in the summer of 1979 in Israel was found by Shelef et aZ.[15] • Their yearly average was 25.2 gm m-2day-1 • The maximum yield wasfOWld at a pond depth of 25 em and without a light filter to remove infra-redirradiance. Yields were reduced by greater pond depth. Further yield valuesare given by various authors in reference 5. With Phaeodaatrflum, Ansell et:al. [16] reported a yield of 8 gm m-2day-l in outdoor culture and Raymond[17] obtained a yield of 41 gm m-2day-1 in an outdoor S-cm thick culture inHawaii. Our yield at the lowest light intensity in batch culture undernutrient-sufficient conditions was 21.74 gm m-2day-l. This value is withinthe range of those reported above. We wonder, if the culture thickness werevaried and the truly optimal light intensity was determined for a given thick­ness, that Phaeodaatylum could not be "pushed" further. To do this we mightalso increase the nutrient concentration in the medium. The nitrogen contentof our present medium would support a crop of 2200 mg liter- 1• It seems nec­essary to obtain a balance between the medium, culture thickness, light inten­sity, and culture dry weight to push Phaecdactulum to its maximum yield. Thesame considerations would be necessary to push any alga and such data wouldbe useful in outdoor culture applications.

The efficiency of light utilization by PhaeodactyZurn in our culture systemvaried greatly with culture conditions (light intensity, N supply, etc.).However our maximum efficiency -- 12.23% utilization of photosyntheticallyactive radiation -- is very close to a value of 13% reported by Raymond [17]for this species in outdoor culture. Goldman [12] suggests that the maximumtheoretical efficiency would be about 20%. Here again, variation of culturethickness 1 cell density, light intensity supplied, and nutrient supply mightincrease the actual measured efficiency.

In fusiform Phaeodaatylum cells Lewin et aZ. [18] reported that the protein,carbohydrate, lipid, and ash contents were 41%, 2%, 34%, and 12% of ~~e dryweight, respectively, and that miscellaneous water- or acid-soluble compoundscontributed an additional 12%. Their culture conditions are not stated. Inhealthy, exponentially-growing FhaeodactyZum cells, Parsons et aZ. [19] re­ported that protein, car.bohydrate, fat, and ash made up 33%, 24%, 6.6%, and7.6% of the dry weight, respectively. Our values for protein (-60%) in N­sufficient cells were much higher than those previously reported. Our carbo­hydrate (-10%) and lipid (-20%) values were intermediate between the previous

55

values~ and our ash values (10-17%) are similar to those reported by Lewinei: at. and somewhat higher than those given by Parsons et: at.

In our N-sufficient cells the sum of the percentages of protein, carbohydrate,lipid, and ash was generally near 100%. Lewin et at. [18] also reported arecovery near 100% but added an unknown water- and acid-soluble fraction tothe protein, carbohydrate, lipid and ash percentages to calculate this re­covery. Parsons et at. I19] obtained a recovery of 73% but did not includesuch a soluble fraction. In our N-deficient cells recoveries were generallylow (60-85%). We did not routinely measure the water-soluble polar materialobtained after methanol-chloroform extraction of lipid samples, but a fewweighings of this material indicated that it was an appreciable fraction. Itmay increase in N-deficient cells~ would increase total recovery~ and wouldbe measured along with the non-polar lipids if we had not washed the methanol­chloroform extract.

We obtained values for calories per carbon of around 10-11 calories mg c- 1•

These are similar to the general value of 11.4 calories mg c- 1 found byPlatt and Irwin [20]. Our values for this relationship did not change withnitrogen deficiency.

Nitrogen deficiency resulted in a large decrease in protein content and anincrease in lipid content. However, N limitation greatly decreased the over­all dry weight yield so that lipid yield was not increased even though cellu­lar lipid content went up. Thus, for this species, we conclude that manipu­lation of cellular N content is not the best procedure to maximize lipidyield.

ACKNOWLEDGMENTS

We are grateful to Mr. Sandor Kaupp for the spectral distribution measure­ments. This work was supported by Contract No. XK-09ll1-l from the SolarEnergy Research Institute.

REFERENCES

1. Wassinck, E.C., Kok, B., and van Oorschot, J.L.P.: The efficiency oflight-energy conversion in Chlorella cultures as compared with higherplants. In Burlew, J.S. (ed.) Algal Culture. From Laboratory to PilotPlant, Carnegie Inst. Washington Publ. No. 600, Washington, D.C., 1953,pp. 56-62.

2. Burlew, J.S. (ed.): Algal Culture. From Laboratory to Pilot Plant,Carnegie Inst. Washington Publ. No. 600, Washington, D.C., 1953, 356 pp.

3. Tamiya, H.: Mass culture of algae. Ann. Rev. Plant Physiol., ~' 309­333 (1957).

56

4. Soeder, C.J. and Binsack, R. (eds.): Microalgae for Food and Feed. AStatus Analysis. Arch. Hydrobiol. Beih. Ergebn. Limnol., Heft 11, 300 pp.(1978).

5. Shelef, G. and Soeder, C.J. (eds.): Algae Biomass. Production and Use,Elsevier/North Holland Biomedical Press, Amsterdam, 1980, 852 pp.

6. Guillard, R.R.L.: Culture of phytoplankton for feeding marine inverte­brates. In Smith, W.L. and Chanley, M.H. (eds.), Culture of MarineInvertebrate Animals, Plenum Publishing Co., N.Y., 1975, pp. 29-60.

7. Strickland, J.D.H. (ed.): The Ecology of the Plankton off La Jolla,California, in the Period April Through September, 1967. Bull. ScrippsInst. Oceanogr., Vol. 17, University of California Press, Berkeley, 1970,103 pp.

8. Sharp, J.B.: Improved analysis for "particulate" organic carbon andnitrogen from seawater. Limnol. Oceanogr., 19, 984-989 (1974).

9. Myklestad, S. and Haug, A.: Produc~ion of carbohydrates by the marinediatom Chaetoaeros affinis var. willei (Gran) Hustedt. I. Effect of theconcentration of nutrients in the culture medium. J. Exp. Mar. Biol.Ecol., ~, 125-136 (1972).

10. Hol~Hansen, O. et al.: Quantitative micro-determination of lipid carbonin microorganisms. Ana1yt. Biochem., 19, 561-568 (1967).

11. Mann, J.E. and Myers, J.:Phaeodaatytum triaornutum.

On pigments, growth and photosynthesis ofJ. Phycol., .~,! 349-355 {1968->.

12. Goldman, J.C.: Outdoor algal mass cultures -- II. Photosynthetic yieldlimitations. Water Research, ~, 119-136 (1979).

13. Goldman, J. C.: Outdoor algal mass cultures -- I. Applications. WaterResearch, 13, 1-19 (1979).

14. Goldman, J.C.: Physiological aspects in algal mass cultures. In Shelef,G. and Soeder, C.J. (eds.) Algae Biomass. Production and Use, Elsevier!North Holland Biomedical Press, Amsterdam, 1980, pp. 343-359.

15. Shelef, G. et al.: Algal mass production as an integral part of a waste­water trea~~ent and reclamation system. In Shelef, G. and Soeder, C.J.(eds.) Algae Biomass. Production and Use~Elsevier!NorthHolland Bio­medical Press, Amsterdam, 1980, pp. 163-189.

16. Ansell, H.D. ei: al.:« Studies on the mass culture of Phaeodaotnflum, II.The growth of Phaeodaotnjlum and ot..l-],er species in outdoor tanks. Limnol.Oceanogr., ~, 184-206 (1963).

57

17. Raymond, L.: Initial investigations of a shallow-layer algal productionsystem. Rept. to Hawaii Natural Energy Inst. and the Dept. of Planningand Economic Development, The State of Hawaii, 27 pp. (1977).

18. Lewin, J.C., Lewin, R.A., and Philpott, D.E.: Observations on Phaeodac­tyZum tricoPnutum. J. Gen. Microbiol., 18, 418-426 (1958).

19. Parsons, T.R., Stephens, K., and Strickland, J.D.H.: On the chemicalcomposition of eleven species of marine phytoplankton. J. Fish. Res.Bd. Canada, 18, 1001-1016 (1961).

20. Platt, T. and Irwin, 5.: Caloric content of phytoplankton. Limno1.Oceanogr., ~' 306-310 (1973).

58

ALGAL RESEARCH AT SERI: A BASIC RESEARCH ON PHOTOBIOLOGICAL

PRODUcrION OF FUELS AND CHEl-lICALS BY MICROALGAE

STEPHEN LIEN

SOLAR ENERGY RESEARCH INSTITUTE (SERI) *

GOLDEN, COLORADO 80401, U.S.A.

*SERI is a Division of MRI operated for the U.S. Department of Energy underCOntract EG-77-C-Ol-4042. The author also acknowledges the vital support forportions of this research from Office of Energy Research, the U.S. Departmentof Energy under OER field task proposal No. 006-80.

59

I. GENE~~ BACKGROUND

The most outstanding feature of the oxygenic photosynthetic organisms

(higher plants, eucaryo tic algae, and cyanobacteria) is their abUi ty to

utilize water molecules as the source of reductant for their biosynthetic

processes. As a result, they are true producers of energy, capable of a net

conversion and storage of solar radiant energy into the chemical free energy

of their biosynthetic products. The algal research project at SERI is a long-

term basic research effort attempting to adapt and to modify (by physiologi-

cal, biochemical, and genetic means) the oxygenic photosynthesis and associat-

ed cellular metabolic processes of the eucaryo tic algae to produce fuels and

chemicals. The basic principle of photobiological fuel and chemical produc-

tion linked to the oxygenic photosynthesis is illustrated in Fig. 1.

(N,P,S. etc.)

+ATP ADP.P.U I

....._-~=----i~ Cell Mass(2)

x ~X H+

I e;'Fd (1) _[H2]~FP H22se

ATP+P. ~NADi+~a~ its I.V"" '\ -

~-;~o~ PSJ ATP \ ~I (PO) I \L __ -J e- hoy ( 3 )rr-. 'cyt1 (CH20)7\rLipids. Oils, &J' t ~PC LHydrocarbons

ADP AT? ,

+ p70a NADPH NADP+Pi +

-0.2

- 0II)-'0>-- 0.2 PSII.....

;UJ

0.4trv/

Figure 1. Oxygenic Phctcsvnt- ests for Fuel Production

60

In the oxygenic photosynthetic organisms, a unique light-driven redox

reaction (which is catalyzed by a chlorophyll-containing, enzyme-pigment

macromolecular complex, known as PHOTOSYSTEM II or PSII, located in the photo­

synthetic membrane systems of these organisms) extracts electrons from water

molecules and liberates molecular oxygen. A second light-driven redox reac­

tion catalyzed by another chlorophyll-containing, macromolecular complex of

enzymes and pigments (known as PHOTOSYSTEM I or PSI) further activates these

electrons generated by the water-splitting reactions of PSII to a redox poten­

tial which is nearly 0.2 V more negative (i.e., having a stronger reducing

power) than that of the H2/H+ redox couple at pH 7. These highly energetic

electrons (reductants) are stabilized in the form of a reduced, low-potential

redox carrier molecule, such as ferredoxin (Fd, a non-heme, iron-sulfur pro­

tein). In the presence of an appropriate hydrogen catalyst (e.g., hydrogen­

ase). the reduced ferredoxin can donate its electron to the hydrogen ions (H+)

yielding molecular hydrogen (H2) , a gaseous fuel [Fig. 1, reaction (1)].

However, under normal photosynthesis, the low potential reductants are utiliz­

ed to generate a reduced nucleotide (NADPH), an universal physiological redox

carrier, which is used for the reductive fixation of carbon dioxide to carbo­

hydrates. The low-potential photosynthetic reductants (such as ferredoxin or

NADPH) also participate directly or indirectly in many reductive biosynthetic

processes leading to the formation of more cell mass as the final product

[Fig. 1, reaction (2)]. Under special conditions, some species of algae can

be induced to channel a large fraction of their photosynthetically fixed

carbon into various b10synthetic pathways leading to lipid biosynthesis [Fig.

1, reaction (3)]. Oils and hydrocarbons may accumulate. In some cases, they

may constitute a large fraction of the total algal mass. Tnese oleaginous

61

algae may lend themselves to future adaptation and development as efficient

solar energy to liquid-fuel transducers.

II. PROJEcr OUTLINES AND PRELL~INARY RESULTS:

Based on the above-mentioned, brief outline of the principle of fuel and

chemical production via oxygenic photosynthesis, the algal research project in

the Photoconversion Branch at SERI emphasizes on the following tlolO areas of

study: (1) hydrogenase system and hydrogen metabolism of green algae; and (2)

lipid metabolism and oil/hydrocarbon accumulation in eucaryotic microalgae.

A. Algal Hydrogenase and Hydrogen Production.

Research on the hydrogenase and hydrogen metabolism of algae is directed

to obtain basic information needed to resolve the major technical problems

(see Fig. 2) associated with algal hydrogen production linked to photobiologi-

cal oxidation of water.

Ferredoxin

//

PseudocyclicFlow/

/

FD-NADPOxidoreductase

8_....-/II\\\

Figure 2. Technical Problem Areas Associated with PhotoproductJon of H2 by AlgaeThe solid arrows indicate reactions leading toward Hz production,whilethe broken arrowsdenote competing side reactions.

62

ClearlYt most of the technical problems arise because whenever water serves as

the substrate for hydrogen productLon, oxygen is liberated as an obligatory

by-product. As oxygen concentration builds UPt various 02-induced back reac­

tions (such as the oxy-hydrogen reaction and autooxidation of the photosynthe­

tic redox carriers) are accelerated. These back reactions reduce the yield of

H2 and consequently reduce the net energy efficiency of the system. More

critically t in systems using algal hydrogenase as the hydrogen catalyst t

oxygen moLecufes , even in very low concentr atLons , rapidly deactivate the

enzyme. Thus t algal photoproduction of H1 from water is generally limited to

very short durations. In most cases, simultaneous 02 and ~ production of

eucaryotic algae lasts only a few minutes. To obtain basic information needed

to solve the technical problems associated with algal hydrogenase, we have

undertaken a systemtic characterization of the hydrogenase system in the

unicellular green alga, c. reinhardti1 t with special emphasis on: (a) under­

standing biochemical events leading to the appearance of active algal hydro­

genase during anaerobic adaptation ~ vitro t and (b) characterizing the bio­

chemical and catalytic properties of the enzyme in vitro.

Evidence was obtained indicating that an energy-requiring (or ATP-consum­

ing) step is involved in the process of hydrogenase activation during anaero­

bic incubation. Recently we observed a very strong stimulatory effect by

various anions on the catalytic activity of the algal enzyme in cell-free

reactions involving the nonphysiological redox carrier, methyl viologen

(MV). Interestingly, we also observed that anions are potent competitive

inhibitors whenever the redox reactions catalyzed by the enzyme is mediated

via the physiological electron carrier, ferredoxin. Additional data and the

technical implication of this anionic effect will be presented.

63

B. Algal Lipid Metabolism and Oil/Hydrogen Accumulation.

The term algal oil and hydrocarbon is defined as that class of carbona­

ceous, lipoidal compounds which are produced by the algal cells, are not

covalently linked to the cellular proteins or carbohydrates, and are charac­

terized by a high content of reduced carbon (such as -CH3, =CHZ' or :Qi)

relative to that of the oxygenated carbon (such as :COH, =C=O or -COOH).

Thus, algal oil and hydrocarbon are, in principle, readily extractable and

easily separated from other cellular consti tuents. Because of their high

reduced-carbon content, algal oil and hydrocarbons have high caloric value and

may be converted into fuels without a large net input of additional chemical

energy.

For the analysis of lipid metabolism and oil/hydrocarbon accumulation in

algae, we have focused our attention, mainly, on unicellular, eucaryotic algal

species of fresh-water and soil origin. The major emphases here are: (1) to

understand the biosynthetic pathways involved in the production of algal oils

and lipids; and (Z) to analyze the control mechanisms involved in the regula­

tion of the oleaginous capacity in algal cells. The ultimate goal of this

research effort is to obtain the basic and fundamental information necessary

for the development of biochemical and genetic manipulations of algal oil and

lipid productivity so that oleaginous algal species can be used as efficient

solar energy to liquid fuel and chemical transducers.

Recently, we developed highly efficient cytochemical staining techniques

for screening and evaluating a large number of algal species for their oleagi­

nous capacity. This staining technique is based on a drastic change in solu­

bility as well as a Large spectral shift accompanying the protonation and

deprotonation of the dye molecule CI Basic Blue 12. In an acidic aqueous

64

solution (pH <5) the dye is protonated and absorbs strongly in the red spec­

tral region. The blue protonated dye in acidic aqueous solution is not read­

ily extracted into the n-nonanl., At neutral to mildly alkaline pH, the dye

readily partitions into the lipoidal phase as the deprotonated free base that

absorbs strongly in the blue spectral region and thereby has a yellowish­

orange color. When samples of algae are pulse stained with the dye ncr Basic

Blue 12" under appropriate conditions, the intracellular oil droplets of the

cells can be readily observed as yellowish-orange structures, while more polar

cellular constituents are stained deep blue (see Fig. 3). Using this techni­

que, a large number of algal species grown under diverse nutritional condi­

tions can be readily screened for their ability to produce and accumulate

oils. Our initial screening identified the following species as good poten­

tial oil or hydrocarbon producers: Neochloris oleoabundans, N. pseudostig­

mata, !:... texensis, and Chlorocroccum oleofaciens. The effect of nutritional

physiological parameters on relative rate of increase of algal mass and oil

content of these organisms is currently being investigated.

65

A. Old cells of C. oleofaciens

accumulate copious amount of

ois (arrows).

B. Oil droplets (arrows) are readily

released from notrogen-starled

cells of N. oleoabundans.\

c. The cells of ~. pseudostig~ata

produce oil of low apparent

viscosity. Under ~oderate ~echanical

pressure the intracytoplasmic oil

(arrows) acc~mulates in t~e space

bet~een the partially ra?tured

cytoplasmic ~embrane and the

intact cell wall.

FIG.3 EXAMPLES OF OLEAGINOUS ALGAE66

DR. JOHN BENEMANN

Ecoenergetics, Inc.

Paper Not Available at Time of Printing

67

ASSESSMENT OF BLUE-GREEN ALGAE IN SUBSTANTIAlLY REDUCINGNITROGEN FERTILIZER REQUIREMENTS FOR BIOMASS FUEL CROPS

D. B. Andersonl, P. M. Molton' and B. Metting2Pacific Northwest Laboratory, Richland, WA 993521

andR&A Plant/Soil, Inc., Pasco, WA 9930,2

ABSTRACT

Laboratory, mass culture and field studies are being undertaken in order to assessthe potential of using blue-green algae (cyanobacteria) as nitrogen biofertilizerson irrigated ground. Of seven candidate strains, two were chosen for applicationto replicated field plots sown to field corn and the basis of laboratory-scalesoil tray experiments at Battelle PNL and ease of semi-continuous 8,000 t cultureat R&A Plant/Soil, Inc. Chosen were Anabaena BM-165, isolated from a local soiland Tolypothrix tenu;s, imported from India. Using the acetylene reduction method,Anabaena is estimated from laboratory soil experiments to be able to fix from30-62 kg N/ha/year, and has been mass cultured to a density of 1527 mg dry wt/t.T. tenuis ;s estimated from laboratory experiments to be able to fix from 27-65 kgN/ha/year, and has been mass cultured to a density of 1630 mg dry wt/z.

INTRODUCTION

Blue-green algae (cyanobacteria) are photosynthetic, O,-evolving prokaryotic micro­organisms, many of which are able to satisfy cellular and metabolic nitrogen needsby fixing N from the air. In addition, certain blue-green algae are able to fixnitrogen unBer aerobic conditions, an ability not possessed by the majority ofother diazotrophic bacteria, due to the presence of specialized N,-fixing cellsknown as heterocysts. For further details regarding the ecology ~nd biology ofblue-green algae, particularly on soil, see Stewart [1] and Metting [2].

It is well known that N2- f i xing blue-green algae are integral components of culti­vated and uncultivated soils [2]. Indeed, free-living blue-green algae have foryears been used in India to supply Nto rice in flooded soils and, to a lesser extent,to irrigated corn, sugarcane and vegetable crops [3]. While it is well documentedthat endemic populations of blue-green algae fix from 15-100 kg N/ha/yr on temperatesoils [4,5,6], the potential for using mass cultures as a nitrogen supplement orsubstitute with irrigated crops in temperate zones has not been assessed.

Information regarding mass cultivation of nitrogen-fixing blue-green algae otherthan that reported by Watanabe [7,8] and Venkataraman [3] is not readily accessible.

OBJECTIVES

The overall goal of the project is to quantify the extent to which the nitrogenrequirements of corn can be met using mass cultured blue-green algae. Specificobjectives include 1) isolation or other procurement of candidate strains ofblue-green algae, 2) selection of the most favorable medium for growth of candidatealgae, 3) monitoring growth and nitrogen fixation of candidate algae on soil inshort term laboratory experiments, 4) quantifying growth of candidate algae in 8,000 2semi-continuous mass culture, 5) selection of two blue-green algae for field testingon the basis of results gathered from laboratory experiments and mass culturetrials, 6) application of mass cultured algae to replicate plots sown to field corn7) quantification of algal growth, nitrogen fixation, fluctuations in soil N, andyield of corn, and 8) assessment of the potential large-scale impact of bio-fertilization

69

with blue-green algae on the overall energy budget associated with production ofirrigated crops.

METHODS AND RESULTS TO DATE

Procurement of Candidate Blue-Green Algae

Objective 1 was accomplished by assembling algal cultures isolated by Metting [9J(Anabaena BM-165, Nostoc muscorum, and CYlindrospermum sp.), and borrowed from G. S.Venkataraman, Indian Culture Collection of Microa1gae (To1ypothrix tenuis, Au10siraferti1issima, Anabaena V-310, and Nostoc v-220). Cultures were maintained on agarslants and in soil-water tubes.

Laboratory Growth Experiments

Experiments were undertaken at Battelle PNL in order to ascertain the best medium inwhich to attempt mass culture of each alga. Media compared were those of Fogg [lOJ,Gorham [11J, Kratz &Meyers [12J, and Watanabe [3,7J. Two hundred and fifty m1 culturesof each of the seven candidate algae were maintained for ten days at 30°C under con­tinuous cool white light at 120 rpm in a New Brunswick Model G-27 PsychrothermIncubator Shaker. Standard inocula were prepared and dry weights obtained as pre­viously described [13]. Experiments were performed in triplicate and repeated twice.

Results showed that Anabaena BM-165 grew best in Gorham's medium, Cy1indrospermum sp.in Kratz &Meyers medium, and Anabaena V-310 in Fogg's medium. The other algae grewequally well in two media, as follows; Nostoc muscorum (Gorham's Kratz &Meyers),Aulosira ferti1issima (Fogg's, Gorham's) To1ypthrix tenuis (Fogg's, Gorham's), andNostoc V-220 (Fogg's, Gorham's).

Soil Tray Experiments

Experiments at Battelle PNL were undertaken using 25 x 50 cm thick-walled pyrex bakingdishes inoculated with late log phase algal cultures homogenized in 15 m1 groundglass tissue grinders and diluted to approximate a field application rate of 65 i/ha.These experiments were undertaken for two reasons; 1) in order that another strongcriterion, i.e., growth on local soil, be tested prior to choosing two blue-greenalgae for field studies, and 2) in order to acquire familiarity with the mechanicsof a) estimating N2-fixation by the acetylene reduction technique [14J, b) estimatinggrowth of algal populations by a ~odi!ied most-probable-number method [15J, andc) monitoring fluxes in soil N03, NH 4 and crop-available N [16J.

Soil was collected from the field site, air-dried, sieved and layered into the dishesto a depth of 5 cm (about 5,000 g). The soil was brought to 125% of field capacitywith deionized water, inoculated with algae using a hand held plant sprayer, andincubated for 26 days at 30°C under cool white light on a 16:8 (light:dark) photoregime.An uninoculated tray was prepared in order to serve as a control with which to studythe native algal flora. Each alga was tested at least twice with visible growthof added algae apparent within ten days in all cases. As inoculation with AnabaenaBM-165 and To1ypothrix tenuis always resulted in even growth on the surface, thesetwo algae were chosen for detailed measurement of acetylene reduction and populationgrowth.

Ethylene production was quantitated with a Perkin-Elmer gas chromatograph by f1ame­ionization detection in one ml samples taken from soil crusts incubated in 10% C2H2for one hr at 30°C under cool white light in 15 ml culture tubes sealed with serumstoppers. Crusts were collected by inserting a one cm diameter cork borer intothe soil, pushing the core through the borer and removing the upper mm with a razorblade. Values reported in Table 1 are averages of three replicates. In no casedid crusts incubated without C2H2 demonstrate values greater than 0.32 ppm C2H4, and

70

was not detectable in most instances. No significant statistical correlations withindexes of soil nitrogen were noted, as was expected in view of the relativelyshort time period and quantities of soil necessary to affect accurate measurements.

Results of the soil tray experiments listed in Table 1 show that both the selectedstrains competed favorably with the native algae and that significantly differentrates of acetylene reduction were measureable. However the extrapolation madeto kg/ha/year of fixed N should only be regarded as an indication of potentialat this point. It is well known that moisture, time of day, the metabolic statusand age of the algal population, and edaphic biological and chemical fluctuationsall influence acetylene reduction [1]; therefore variation in the values listedare not surprising but rather expected.

Table 1. C H -reduction by Algal Crusts and Extrapolation toLarge-Scale N~-fixation for Anabaena BM-165, Tolypothrix tenuis and

Native Algae.

Algal FloraAnabaena BM-165

Tolypothrix tenuis

Native

Ageppm CzH4/ 44mm2/ hr MPNT(days) kg/ha/year

55 6.80 47-62 2.0 x lag12 5.71 39-52 2.4 x 10919 4.40 30-52 1.0 x 10926 3.83 26-35 5.2 x 1065 7.22 49-65 1.9 x 10912 6.86 47-62 1•1 x 1°919 6.13 42-55 6.7 x 10926 3.99 27-36 4.8 x 1045 0.79 5-7 2.0 x 105

12 2.8 x 10519 7.7 x 10~

26 3.09 21-28 5.1 x 10/

*Range of values represent limits of theoretical ratios of acetylene reduction tonitrogen fixation. Where one year ;s equal to a 100 day season of 10 hour days.

tMost-Probable-Number of blue-green algal particles per gram air-dry so;l.

Mass Culture

Semi-continuous 8,000 t mass culture experiments at R&A Plant/Soil, Inc., wereundertaken to detenmine ease of preparation of inocula for field tests. Each ofthe seven candidate algae were mass cultured at least twice in Gorham's medium. OnlyAnabaena BM-165 and To1ypothrix tenu;s were successfully cultured to a maximum densityof at least 1000 mg/t and less than 20% contamination (as assessed by cell counts)with unwanted algae and protozoa. Average values for final density, specific growthrate and mean doubling time are included in Table 2.

Table 2. Growth Parameters for Mass Cultured Algae.

Alga Specific growthrate j..I

Mean doublingtime(hr)

Maximumdensity

(mg/2.)

Anabaena BM-165Tolypothrix tenuis

.,,../0

.26

71

2443

17001630

June 2

April 17

May 1-3May 20May 27-28May 29

Field Studies

On the basis of soil tray and mass culture experiments, Anabaena BM-165 andTo1ypothrix tenuis were chosen as the two algae to be used in the field work.

Fifteen 15 m2 plots were established, including three each inoculated with one orthe other selected alga, three fertilized with 50 kg N (as NH 4N01)/ha, three with100 kg N/ha and three without algae or fertilizer. In Table 3 are summarized thoseevents relevant to field work through June 2.

Table 3. Chronology of Events Associated with FieldStudies of Bioferti1ization with Blue-Green Algae (1981).

Alfalfa sprayed with RoundupR (isopropyl aminesalt of N-phosphomethy1 glycine)Seedbed prepared.Northrup-King PX-14 field corn sown.Corn emerges.Anabaena BM-165 inoculated onto plots at about380 t/ha of a 1527 mg/t mass culture.To1ypothrix tenuis inoculated onto plots atabout 380 t/ha of a 1630 mg/t mass culture.

Literature Cited

1. Stewart, W. D. P. Systems involving blue-green algae (cyanobacteria). inMethods for Eva1uatin 8io10 ical Nitro en Fixation. F. J. Bergersen (ed.).Wi ey &Sons, New York 1980, pp. 583-635.

2. Metting. B. The systematics and ecology of soil algae. Botanical Review 47 (2), inpress (1981).

3. Venkataraman, G. S. Algal Biofertilizers and Rice Cultivation. Today andTomorrow's Printers and Publishers, New Deh1i (1972).

4.

5. Dart, P. H. and J. M. Day. Non-symbiotic nitrogen fixation in soil. inSoil Microbiology. N. Walker (ed.). Wiley &Sons, New York (1975), pp. 225-252.

6. Henriksson, E. A., L. E. Henrikssen and E. J. DaSilva. A comparison of nitrogenfixation by algae of temperate and tropical soils. in Nitrogen Fixation byFree-Livin Microor anisms. W. D. P. Stewart (ed.) Cambridge Univ. Press,Cambridge 1975, pp. 199-206.

7. Watanabe, A. On the mass-culturing of the nitrogen-fixing blue-green algaTo1ypothrix tenuis. ~. Gen. Microbio1. 5, 85-91 (1959).

8. Watanabe, A. Collection and cultivation of nitrogen-fixing blue-green algaeand their effect on the growth and crop yield of rice plants. Pro. Symo.A1go1ogy, New Dehli (1960).

9. Metting. B. A Comparative Study of Algal Communities on Cultivated andUncultivated Portions of a Schumacher Silt Loam. Ph.D. Dissertation, WashingtonState Univ., Pullman (1979).

72

10. G. S. Venkataraman, personal communication.

11. Gorham, P. R., J. S. McLachlan, U. T. Hammer and W. K. Kim. Isolation andculture of toxic strains of Anabaena flos-aauae (Lyngb.) de Breb. Verh. Int.Var. Limnol. 15,796-803 (1964). ----------

12. Starr, R. C. The culture collection of algae at the University of Texasat Austin. ~ Phycol. 14, supplement (1979).

13. Metting, B. New species of green microalgae (Chlorophycophyta) from anEastern Washington silt loam. Phycologia 19, 296-306 (1980).

14. Turner, G. L. and A. H. Gibson. Measurement of nitrogen fixation by indirectmeans. in Methods for Evaluating Biological Nitrogen Fixation. F. J.Bergersen (ad.). Wiley &Sons, New York (1980), pp. 111-138.

15. Zimmerman, W., B. Metting and W. R. Rayburn. The occureence of blue-greenalgae in silt loarns of Whitman County, Washington. Soil Science 130, 11-18 (1980).

16. Bremner, J. M. Nitrogen availability indexes. in Methods of Soil Analysis.C. A. Black (ed.). Amer. Soc. Agron., Madison (1965), pp. 1324-1345.

73

WATER HYACINTH WASTEWATER

TREATMENT SYsrEM

C. A. Lee, M.S.Assistant Project Manager

WED EnterprisesGlendale, California 91201

T. McKim, M.S.Project Engineer

Reedy Creek Utility CompanyLake Buena Vista, Florida 32830

ABSTRACT

A prototype water hyacinth wastewater treatment system has beenin operation at Walt Disney World, Florida, since July of 1979.During this period, this system has been operating with threeobjectives: 1) to demonstrate an unconventional energyconservative wastewater treatment system; 2) to determine theability of water hyacinths to treat primary effluent to secondarytreatment standards in a flow-through system; and, 3) to try tomaximize production of a potential energy feedstock whiletreating wastewater. Calculations indicate the water hyacinthsystem requires less than 50% of the energy needed to run acomparably sized conventional secondary treatment system. Theeffluent from the water hyacinth system demonstrates 80-90%removal of total suspended solids and B.O.D., meeting secondarytreatment standards when coupled with primary treatment. In thefirst year of operation, the amount of water hyacinths harvestedfrom the 3/4 acre system was approximately equivalent to 21 drytons/acre-year (47.8 dry metric tons/hectare-year). The effectsof varying harvest routines and detention times are beinginvestigated to try to opt~ize wastewater treatment and biomassproduction. Studies on nutrient removal as well as biomassconversion to methane gas are also being performed.

75

In Phasewill add

4)

WATER HYACINTH WASTEWATER

TREATMENT SYSTEM

C. A. Lee, M.S.Assistant Project Manager

WED EnterprisesGlendale, California 91201

T. McKim, M.S.Project Engineer

Reedy Creek Utility CompanyLake Buena Vista, Florida 32830

INTRODUCTION

The Water Hyacinth Wastewater Treatment System at Walt DisneyWorld has been in operation since July 1979. The project isoperating with these objectives.

1) To determine an unconventional energy conservativewastewater treatment system.

2) To determine the ability of water hyacinths to treatwastewater to both secondary and tertiary treatmentstandards in a flow through system.

3) To maximize production of a potential energy feedstockwhile treating wastewater.

II of this project, part of which has already begun, wetwo more objectives to our operational goals, they are:

to demonstrate and optimize the bioconversion of waterhyacinths to methane.

5) to treat wastewater to tertiary treatment standards.IGT through sponsorship with GRI has performed the baselinestudies on methane production from Walt Disney World waterhyacinths, and operation of two 50-liter digestors, designed byIGT, is already underway at Walt Disney World.

The SERI Contract will expand the system by intergrating three(3) additional channels with the same dimensions, materials anddesign of the orginal channels.

The purpose of the addition will allow for more rapid evaluationand optimization of the systems parameters, maximize plantproductivity, and increase the biomass production capability.

76

The objective will be to maximize plant productivity withoutadversly affecting the systems capacity to treat wastewatere ffec tive ly.

With the SERI addition to the system, the invest ig~tions willinclude evaluation of the following variables on water qualityand plant productivity:

1) Increased detention times2) Use of a protective cover during winter months3) Use of mechanical aeration4) varying harvesting routines5) Addition of nutrients and/or met~ls for max1m1z1ng

phosphorus removal and maximizing growth.

DESIGN OF SYSTEM - PROJECT COMPONENTS

The system can be broken down into several components: 1) theProduction/Treatment channels; 2) the Harvesting System; 3) theCompost System. Special studies are also being performed.

Production/Treatment Channel:

Components of the production system include three 1/4 acreconcrete channels, the system piping, utility tie-ins, hydrauliccontrol and metering devices, and pumping stations.The walls of the channels are constructed of reinforced concreteblocks on a cast-in-place, reinforced concrete foundation. Thechannels (29' x 360') are lined with 20-mil PVC tacked to the topof the channel walls with lumber. PVC booms are tied off tocleats along the top of the channel walls every sixty-feet andact as a corral preventing the hyacinths from packing together atone end. See Figure #1. The water level in the channels isadjustable, and can be maintained at depths of 14 to 36 inches.The influent gate valves incorporate the capability for eachchannel to be operated independently and at varied depth. Thesystem is currently treating 63,000 gpd but the system ishydraulically designed to handle flows up to 200,000 gpd. Theinfluent flow is split, with 25,000 gpd to Channell, 21,000 gpdto Channel 2 and 17,000 gpd to Channel 3.Two submersible pumps, one in the effluent channel of the primaryclarifier and the other in a filter pump wetwell of the existingReID Wastewater Treatment Plant can provide the channels withprimary and/or secondary effluent. The channels are alsointerconnected hydraulically to provide a variety of experimentalmodes of operation (See Figure #2).

77

Harvesting

The equipment used in harvesting include: 1) a front end loader;2) a double belt conveyer-chopper; and, 3) a forage wagon. Allmechanisms in the system are powered hydraulically from the frontend loader. The harvesting system is designed with a capacity of50 tons/hour, far in excess of the system's requirements.Harvesting is accomplished by physically pushing the hyacinthsonto the primary conveyor with a long handled hook. At the endof the primary conveyor, a flail chopper coarsely cuts the plantswhich are then conveyed to the forage wagon. The forage wagonhas a live bed for ease of loading and unloading.

The three channels are divided into six cells 60' long by 29'wide, by floating PVC booms. The booms can be used to push thehyacinths into a uniform density. thereby allowing for a densitymeasurement before and after harvesting. This method assumesthat a uniform density is achieved through compaction of thehyacinths. however density data collected in this mannerindicated this method was somewhat inconsistent. At the presenttime. another method of density measurement is being used. Inthis method, one meter square trays are stocked with a targetweight of hyacinths. Each week the tray is weighed and onceevery other week a portion of the cell is harvested to maintainthe target restocking weight. The one meter square traytherefore is used as a representation of the growth of the entirecell. Both methods have often demonstrated crop reductionwithout the benefit of harvesting. It is believed that the waterhyacinths under dense conditions (over 4 wet lbs./sq. ft.) startto slough off material and in many cases the crop weight issignificantly reduced.

Composting

The composting operation utilizes a windrow system with a compostpad and front end loader. Once the forage wagon is full. it ismoved to the compost pad and unloaded by means of the live bed.All hyacinths harvested from one channel are put into a pile. Atthe beginning of system operations a new compost pile was turnedthree times/week during the first week and once a week in thefollowing weeks. Measurements on temperature and free moistureindicated that the desired values of 50-60% moisture and140o-l50oF (1] were not reached in the compost piles. It wasthen decided to stop turning the piles. The unturned pilesreached a maximum temperat~re of 12loF, but dropped to 80 0F

after two days.

80

Arter a 4-6 month period compost piles resemble a potting soilwith an 80+% volume reduction requiring little or no operatorattention.

Special Studies

Trace metal studies on Walt Disney World water hyacinths havebeen performed by the University of Arizona EnvironmentalResearch Lab. Initial studies indicated that although WaltDisney World water hyacinths were extremely rich in iron, theycontained less Ca, Mg, and K than did water hyacinths grown innutrient solutions. See Table 1. The uptake of these cationsmay be inhibited by the uptake of NH4 + [2]. However, tissuestudies performed later in the year did not indicate a lack inCa, Mg, K. Iron has already proven to be a necessary element inpreventing chlorosis (yellowing of the plant), [3, 4]. However,Walt Disney World hyacinths appear to take up an abundance ofiron. The large amount of iron found in the tissue may beinhibiting growth and additional studies are being made by theuniversity of Arizona. Further studies will be done to determinethe ability of trace metal uptake versus the availability of themetals in Walt Disney World primary effluent.IGT throughsponsorship with GRl is performing the baseline studies onmethane produc tion from wnw water hyacinths (See Figure fF3). Theinitial study indicates an average yield of 3.0 scf ofmethane/lb. of volatile solids added to the digestor from WaltDisney World hyacinths. Methane studies will continue as two50-liter digestors are now being operated at wnw in preparationfor a larger digestor or Process Development Unit that willeventually utilize all of the water hyacinths harvested from thethree channels.

OPERATIONAL MODES

The water hyacinth wastewater treatment system operation began onJune 18, 1979 utilizing secondary effluent, and a parallel flowmode. All three channels were seeded with water hyacinthscovering approximately 10% of the surface area. Within twomonths there was essentially 100% plant coverage. The targetflow rate was set at 50,000 gpd or approximately 17,000gpd/channel. On July 13, 1979 at the request of EPA the channelswere switched to treat primary effluent. The channel depths werevaried with Channel fFl at 14" and Channels {;2 and iF3 at 36". Atthese depths, the detention time for Channel #1 was 5.36 days,and 13.8 for Channels #2 and #3. Water quality measurementsbegan on July 2, 1979 and followed the schedule in Table #2.

81

TREAT~ENT SYSTE~

TABLE 2

~·10NITORING

A. Water Qualitv - Influent & Effluent, Each Channel

1) Daily Tests

a) pHb) D.O.c) water temperatured) Insolation - Totale) Rainfall - Totalf) Air temperature and relative humidityg) Influent and Effluent Flow - Totals

2) Twice per Week

a)b)c)

d)e)f)g)h)

TSSTSNH+-N

40- P04T- P04N03-NBODSTKN

3) Twice per Month

a) Total and Fecal Coli formsb) Chloridesc) Alkalinity

B. Bio Mass Production - Each 60' Cell

1) Density (lbs/ft2) - 1 ?er week

~able 2 - ~oni:orinq roueine

84

Harvesting began on August 2nd with a harvest schedule set attwice a week for Channels #1 and #2, while Channel #3 washarvested twice a month. At this time, the compaction method ofmeasuring density was utilized to determine the percentage ofsurface area covered by the crop. A five square foot "cookiecutter" would then segregate water hyacinths for weighing andfrom this weignt and the percentage of covere~ surface area, theweight of the entire crop was established.

Subsequent to the first water hyacinth review meeting, held onOCtober 19, 1979, including all participants in the project, theoperational procedures were changed to the following:

1) Channels 4~1 and iF2 were both maintained at a depth of 1411

receiving primary effluent.2) Channel #1 was harvested twice a week from sequential

cells throughout the channel.3) Channel #2 was harvested once a month and only from the

influent end.4) Channel #3 was used as a control channel maintained at

14" with no water hyacinths.

In January, 1980, the same harvesting schedule was main- tained,however, a seed crop density of two wet lbs./sq. ft. was left ineach channel to reseed the harvested area. During this time, PVCbooms were used to compact the crop for density measurements.Following the second water hyacinth review meeting, held on April3, 1980, one major operational change was made:

1) Channel #3 was restocked with water hyacinths and thatchannel used for density-yield studies. These studiesincluded use of twelve, one-meter square trays made ofPVC piping and plastic vexar mesh. These trays werestocked at target densities ranging from 0.5-3.0 Ibs./sq. ft. with duplicates for each target density. Thesetrays were weighed twice a week and the amount over thetarget density was harvested. At the end of six weeksthe optimum density for restocking proved to be 1.5 wetlbs./sq. ft. See Figure #4.

From the above method of measuring growth, it was hoped that thetray method would prove more consistent as a measure of cropdensity and weight, and the method would be used in the other twochannels. At this time it is not clear which method is superiorfor measuring crop yield.

85

At the next review meeting, held on November 11, 1980, thefollowing operational decision was made:

1) Increase flow rates in channels one and two byapproximately 50% and 25% respectively.

In November, breakdown of the harvesting equipment preventedharvesting during the months of NOvember and December. Coldweather also prevented harvest in January. In spite of thesehardships and the increase in flow to channels #1 and #2, thesystem still managed to remove 90% TSS, however, BOD and nutrientremoval did decline.

In March of 1981, two SO-liter digestors were seeded and gascollection and analysis initiated at Walt Disney World. Thepurpose of this work is to determine the effects on gas qualityand quantity from any seasonal changes in the feedstock of waterhyacinths and primary wastewater sludge.Currently one digestor isfed primary wastewater sludge and the other is operated on waterhyacinths. The digestors are batch fed each day at a rate of0.15 lbs. volatile solid/cubic ft. of digestor volume. Both areoperating on a 15 day detention time. At the time of writing,the digestors have not yet reached steady state.

At the next review meeting held on April 28, 1981, the followingoperational decisions were made:

1) Channel #3 would be harvested once/week at the influentend only and would maintain a restocking density of 1.5wet lbs./sq. ft.

2) Channels #1 and #2 would also use a restocking density of1.5 wet lbs./sq. ft with a harvesting interval oftwice/month. Harvest will occur at the influent end only.

3) The flow of 63,000 gpd would be maintained for 3additional months with the regular harvest routine. Atthe end of 3 months, the most efficient and effectiveflow rate for water quality would be evaluated andchannels #1 & #2 would then be used in series at thatchosen flow rate.

87

RESULTS AND DISCUSSION

Water Quality

The water hyacinth system has proven to be an effective secondarytreatment system. Coupled with primary treatment, the system hasmet the State of Florida's secondary treatment levels of 90%removal of TSS and BOD. see Figures #5 & #6. The operationalchanges can be seen in this BOD and TSS profile spanning July1979 - March 1981. From July, 1979 to October 1979, Channel #2and iF3 were operated at a 36" depth and the water quality dataindicate that the deeper depth was not quite as effective as the14" depth of Channel tn. In October, 1979, Channel/F3 wasoperated as a control channel with no water hyacinths. Thegraphs indicate a marked increase in suspended solids at thistime due mostly to algal generation and a decrease in BODremoval. Also at this time, Channel #2's depth was decreasedfrom 36" to 14" and a corresponding improvement in the effluentof Channel #2 is observed. In May, 1980, when Channel #3 was online once more stocked with water hyacinths, a marked improvementwas observed in the water quality of Channel #3's effluent.

In the summer months of June, July and August, 1980, a decreasein water treatment is observed in both Channels #1 and #2. Atthis time, many plants were beginning to wither and die. Atfirst, the unhealthiness was attributed to the excessive heat anddryness of an unusual Florida summer, where temperatures reachedover lOOoF for over three consecutive weeks without anyafternoon showers. However, by July, close examination of theplants showed that a moth, Sameodes albiguttales, had infestedChannels #1 and #2. In its caterpillar stage, the insect boresinto the stolen of the plant cutting off the transport ofnutrients from the roots to the leaves. Other insects were alsofound in the channels including the weevil, Neochetinaeichhorniae which eats the leaves in banded patternsapproximately l-2mm thick. (5], another species of weevil N.bruchi and the spider mite, Bryobia praetiosa. The insects­appear to attack more heavily when the water hyacinths arestressed, the weevils in cases of density stress [5] and the mothin cases of hot, dry weather stress [6]. The spider mite andweevils are found year round, however, spot harvesting appears tocontrol them very well. It is believed the moth caused the mostdamage to the channels this summer. Channel #3, restocked in Maywith water hyacinths, was the only channel not badly attacked.

88

In August, all channels were treated with the insecticide Sevin,to prevent further spread of the insects. By the beginning ofseptember, Channels #1 and 12 were beginning to recover and bythe end of September they had completely recovered, with BOD andTSS removal returning tO,higher levels. The application of Sevinwas stopped on September 5. Although the hyacinths were able torecover, it was impossible to determine if the insecticide Sevinor if the break in the hot dry spell was responsible for recovery.

The effluent water quality from the channels is very goodespecially when the effect of evapotranspiration is considered.In October of 1980, a meter and totalizer were installed at theinfluent end to monitor the volume of water going into thesystem. The water budget data from the totalizer and theeffluent meter indicate a 12-26% evapotranspiration rate and thisfinding is consistent with previous studies [2]. See figure 7.A water budget is extremely important in evaluating water qualitytreatment and efforts are being made to monitor the water volumein individual channels rather than measuring the three channelstogether.

In November and December of 1980, equipment failure preventedharvesting and in January 1981, extremely cold temper~tures frostburned much of the crop. Recovery from the frost period was veryrapid. The water hyacinths were green within two weeks andrecovery was complete within one month. Harvesting resumed inMarch. In spite of the equipment failure and lack of harvesting,the cold weather, and increased flow to channels one and two, thesystem still managed to remove 80-90% of the suspended solids.The removal of BOD was not as effective during this period,averaging between 55-77% removal. However at this time it isdifficult to assess how much of the reduced performance in BOD iscaused by the colder temperatures, the lack of harvesting, or theincrease in flow.

Water quality declined in Channel 3 which was restocked in May1980 with water hyacinths and not ha=vested until October 1980.During that six month period, water quality improved during thefirst two months and declined in July and August. Although notas badly attacked as channels 1 and 2, Channel 3 never recoveredfrom the insect infestion and the lack of harvesting appears tobe responsible for the continued decline in water quality ofChannel 3.

91

Nutrient removal in the channels, although fairly consistentthroughout the year, does not reach the 90% removal range of TSSor BOD. The channels appear to average approximately 30-50%removal of TKN and 20-40% removal of TP04. See figures 8 & 9.

Healthy plant growth is very important for water ~uality,

particularly in nutrient removal as is demonstrated by the graphsof TP04 and Total Kjeldahl Nitrogen removal (see figures 8 &9). The months of January and February in 1980, January ~nd

February in 1981, and July and August 1980 are the months showingthe least reduction in nutrients. In each case, the waterhyacinths were under stress, in January and February by coldtemperatures inhibiting growth, and in July and August by theinsect infestations which also prevented healthy growth.Harvesting out some of the dead plant material which releasednutrients would probably have improved the water quality duringthe periods of stress, unfortunately due to equipment failureharvesting was not possible during January and February of 1981.

Nutrients appear to be removed at a rate proportional to theloading of the system, even during the winter when the growthrate of the water hyacinths is slower. Bacteria continue toremove nitrogen in the colder months, however, during the warmermonths the hyacinths appear more responsible for nitrogenremoval •. See Figure 10. The bar chart indicates the totalamount of nitrogen removed from channel one and the amountremoved by the water hyacinths. This chart indicates a trendshowing the hyacinths playing a more important role in nitrogenremoval during the warmer months.

Phosphorus removal also follows a similar trend to total nitrogenremoval in that more phosphorus is removed from the system thancan be accounted for in the plants. See figure 11. The amountof nitrogen and phosphorus that is removed in excess of what isfound in the plants indicates that there may be some depositionof sludge at the bottom of the channels however no accumulationof sludge is observed. One explanantion may be that as thesludge degasses it floats to the top and is removed duringharvesting operations.

93

In January 1981. the first set of trace metals were analyzed fromthe influent and effluent ends of each channel. Data wascollected for the following trace metals: boron, calcium.magnesium, potassium, copper, iron, manganese, zinc, lead,cadmium, chromium and arsenic. See Table 3. The systemdemonstrated the greatest percent removal of lead (60-78%removal) while potassium showed the least percent reduction(1-4%). This system seems slightly more efficient than anotherstudy [7] using a continuous flow through system for removingboron while showing less efficiency at removing arsenic.Unfortunately due to the frost in January 1981, no tissue sampleswere taken, however tissue samples taken in the first quarter of1980 indicated an abundance of iron in the water hyacinths. SeeTable 1. This abundance of iron appears to be normal for waterhyacinths grown in wastewaters [8]. Disney water hyacinths alsocontain more zinc and less manganese than water hyacinths growingin natural stands [9], wastewaters [8], and nutrient solutions(see Table 1). Trace metal samples from the influent, effluentand tissues will continue on a once/quarter basis throughout theremainder of this study.

Productivity

From January 1980 - June 1980 Channels #1 and #2 wereconsistently harvested in the following manner: In Channel #1the cells were sequentially harvested once a month while Channel#2 was harvested from the influent end only once a month.Channel #1 produced the e~uivalent of approximately 21.3 drytons/acre-year (47.7 metric tons/hectare-year); Channel #2produced 20.7 dry tons/acre-year (46.4 metrictons!hectare-year). Unfortunately the channels could not beharvested during the months of July and August, because of themoth infestation. Looking at the harvesting profile there doesnot seem to be any distinct advantage in one harvest pattern overthe other in terms of productivity, one channel doing better onemonth, the other the next. See Fig. 12. The decline in growthin June, 1980, observed in Channel #2 is most likely explained bythe moth infestation. Channel 3 which was restocked in May 1980and not harvested until October 1980, produced only 16.20 wettons during that period while Channel 1 produced 35 wet tons andChannel 2 produced 31 wet tons in that same six month time span.The difference in yield clearly indicates the importance ofcontinued harvesting for maximum productivity.

98

TABLE 3

AWRAGE VALUES FOR JANUARY 19 Sl

Influent il i2 43 % Removal

11 12 i3

Boron mg/l 0.14 0.10 0.09 0.13 38.2 46.4 26.0

Calcium mg/l 33.6 27.1 34.3 35.5 30.5 14.7 15.8

Magnesium mg/1 5.6 6.1 6.1 6.1 6.2 39.4 13.4

Potassium ~g/l 9.7 10.8 11.6 12.4 4.1 0 (2.0)

copper ppb 27.9 27.9 20.8 31.4 13.S 36.4 11.0

Iron ppb 457.8 364.8 377.2 347.9 31.3 31.1 39.4

Manganese ppb 18.2 13.1 12.8 15.6 37.6 40.0 33.3

Zinc ppb 280.8 407.5 388.7 444.9 (24. 7) (15.4) (26.2)

Leaa ppb

Cadmium ppb

Chromium ppb

Arsenic ppb

12.8

0.35

0.83

0.8S

4.9

0.17

0.72

0.88

1.1

0.95

0.88

3.5 66.7 60.0

0.31 66. 7 (133)

0.72 25.0 14.3

0.88 12.5 14.3

78.0

25.0

28.0

26.7

Table 3 - ~verage tra~e metal eoncentracions sampled durin9 January1981 .

99

Productivity and nutrient uptake are certainly related, howeverin a system such as this, with virtually unlimited nutrients forthe plants, the growth rate of the hyacinths is mostly controlledby the environment particularly ambient temperatures. Seefigures 13, 14, 15. These graphs plot the monthly yield versusthe average monthly ambient temperature, insolation, and watertemperature. From these graphs the ambient temperature has thehighest correlation coefficient with yield. A protective coverduring the winter months should increase the yield of waterhyacinths. In addition a cover provides the possibility ofutilizing C02 enrichment, which according to studies at theUniversity of Arizona can stimulate growth up to four times. [10]

Previous reports have indicated that water hyacinths can produce40-60 dry tons/acre-year (89.6 - 134.4 dry metrictons/hectare-year) [2,11, 12]. In addition to varying theharvest routines, and the restocking densities to optimizegrowth, the possibility that there may be something in WaltDisney World primary effluent that may be inhibiting or limiting~owth is also under investigation. Future laboratory studies byUniversity of Arizona under controlled conditions will indicateif there is growth inhibitor or a limiting element in the primaryeffluent.

Energy

Preliminary calculations indicate that the water hyacinth systemwould utilize less than 50% of the energy required for aconventional secondary treatment system utilizing activatedsludge. See Table #4. These tables are average monthly energyrequirements for the Water Hyacinth Systm and the Reedy CreekUtility Co. secondary treatment system scaled to one M.G.D.These are energy figures which do not consider the amount ofenergy produced by the conversion of water hyacinths to methane.Using 3.0 scf methane/lb. of volatile solids, [13] The waterhyacinth system at 25 dry tons/acre-year (56.0 dry metrictons!hectare-year) would produce 1.4 x l08BTU's/month permillion gallons of wastewater treated each day. Thesepreliminary calculations, based on biomass yields for one year,indicate that the water hyacinths could produce enough methane tooperate the channels.

101

WATER HYACINTH WASTEWATER

TREATMENT SYSTEM

TABLE 4

Energy requirements for the Water Hyacinth Wastewater Treatment

System at 50,000 gpd for 3/4 Acre System (as a secondary treatment).

Influent pump 1.8 x 10 6 BTus/month

Effluent pump 1.2 x 10 6 BTUs/month

Harvester 3.7 x 10 6 BTus/month

Forage Truck 3.5 x 10 6 BTus/month

10.2 x 10 6 BTus/month

AT 1 MGD using 15 acres (estimated) 1.30 x 10 8 BTus/month

ReID Wastewater Treatment Plant energy requirements for secondary

treatment at 3.3 MGD

Aeration Tanks

Clarifiers

AT 1 MGO (estimated)

9.2 x 10 8 BTus/month

1.4 x 10 6 BTus/month

9.21 x 10 8 ETus/month

2.7 x 10 8 BTus/month

:'abh 4 - ~n~r?y requirements ~or th~ ~acer hyacinth ~aste~a:e:trea~~ent s:~~em· !nd the ~eedv C:e4K r~orovemen~

O~st~ict ~as~ewater Trea~~ent·?lant. .

105

FUTURE OPERATIONS AND MILESTONE CHART

In addition to the 3 channel expansion of the water hyacinthwastewater treatment system, the future operations also plan foran advanced anaerobic dige stion process for converting most ofthe water hyacinths to methane gas. The Process Development Unit(PDU) is presently being designed by IGT under sponsorship ofGRI. Installation and Operation of the PDU is scheduled for July1982.

The revised milestone chart for the execution of the SERIcontract is shown in Figure 16.

CONCLUSION

The Water Hyacinth System at Walt Disney World has demonstratedit does obtain secondary wastewater treatment standards. Thefirst year's operation has produced good water quality andyielded the equivalent of 21 dry tons/acre-year (47.8 dry metrictons/hectare). The preliminary calculations based on one year'soperation, indicate that the water hyacinth system not onlyutilizes 50% less energy than a comparable sized conventionalsecondary treatment system, but by converting water hyacinths tomethane, the treatment system could be energy self-sufficient.

106

REFERENCES CITED

1. Finstein, M., "Microbial Ecosystem for Composting".Paper presented at 10th Annual Composting and WasteRecycling Conference, Compost Science/Land UtilizationConference, Los Angeles, Ca., May 5-7, 1980.

2. Ryther, J. H., "Cultivation of Macroscopic MarineAlgae and Freshwater Aquatic Weeds," Progress Reportprepared for the Solar Energy Research InstituteContract No. XR-9-8l33-l, January 1980.

3. Phillips, J. M. and Hartung, R. F., "Solar Energy forLiquid and Gaseous Fuel Production by Means ofBioconversion," Final Report prepared for the ArizonaSolar Energy Research Commission State Contract No.OEPAO - 368-77 (AG C2l30), 1977.

4. Stewart, E. A., IIUtilization of Water Hyacinths forControl of Nutrients in Domestic Wastewater Lakeland,Florida." Progress Report to Dawkins and Assoc , ,Orlando, Fla., Sept. 1979.

5. Swett, D. PhD., and Padera, C. Research DevelopmentManager, Per Comm. Coral Ridge Properties, Division ofWestinghouse, Coral Springs, Fla., Aug. 8, 1980.

6. Quimby, Jr., P. C. "Biological Control TechnologyDevelopment," Proceedings, Research PlanningConference on the Aquatic Plant Control Program, NewOrleans, LA. Miscellaneus Paper A-7a-l, Office, Chiefof Engineers, U.S. Army. Oct. 3-6, 1977.

7. Tridech, S., A. J. Englande, Jr., Michael Hebert, Jr.,R. F. Wilkinson "Kinetics of Trace Contaminant RemovalFrom Secondary Domestic Effluent by Vascular AquaticPlant Systems" Paper presented at the Water PollutionControl Federation Conference, Las Vegas, Nevada,October 1, 1980.

8. Wolverton,Hyacinth:~Vol.

B. C. and McDonald, R. C. "The WaterFrom Prolific Pest to Potential Provider"8, No.1 p , 2-9,1979.

108

9. Boyd, Claude E. "Vascular Aquatic Plants for MineralNutrien t Removal from Polluted Wa ters It Economic Betany24: 95-103, 1970

10. Fontes, M. PH.D. Research Horticulturist, Per. Comm.,Environmental Research Laboratory, University ofArizona, Tuscan, November 13, 1979.

11. Benemann, J.R., "Energy from Aquaculture BiomassSys tems: Fresh and Brack ish Water Aquatic Plants, II

Final Report to Office of Technology Assessment U.S.Congress, Task X, Oct. 10, 1979.

12. Wolverton, B. C., and McDonald, R. C., "Water Hyacinth(Eichhornia crassipes) Productivity and HarvestingStudies lt section I Part 3 ERL Report No. 171TMX-72729, NASA, NSTL Bay St. Louis, Mo. Mar 1967.

13. Chynoweth, D. P. and Ghosh,S., "Energy Conversion ofSeTN'age Sludge and Water Hyacinth" Report presented atProgram Review Meeting of the Water HyacinthWastewater Treatment System Walt Disney World, Fla.No',. 11, 1980.

109

CULTIVATION OF MACROSCOPIC MARINE ALGAE

J. H. RytherWoods Hole Oceanographic Institution

Woods Hole, Massachusetts 02543

ABSTRACT

High yields of Gracilaria were obtained in cultures suspended by aerationwith rapid exchange of enriched seawater. Attempts were made to reduce thehigh cost inputs of aeration, seawater flow and nutrients and still main­tain high yields. Physiological research was required to understand thefunction of those inputs and how to reduce them. Gracilaria soaked innutrient solutions for several hours was capable of storing sufficientnutrients for subsequent growth for one-two weeks. Intermittant aerationfor a few hours/day was equally effective as continuous aeration for growthenhancement. Rapid seawater exchange is necessary to maintain pH control,since Gracilaria has little ability to utilize bicarbonate and virtually nofree C02 exists at pH 9.0, commonly reached in high-yield, low seawaterexchange cultures. Ulva and other species that use bicarbonate readily maybe better marine biomass candidates than those like Gracilaria and Macrocyst~

that lack that ability.

INTRODUCTION

The highest yield~ of marine biomass that have yet been reported to date on asustained basis are those of the red seaweed Gracilaria tikvahiae. Grown insuspended culture by vigorous aeration and with 25-vo1ume exchanges per day(retention time 0.04 days) of enriched seawater, the seaweed produced theequivalent of 129 dry tons per hectare per year, grown throughout the year in50 1 outdoor cultures in Fort Pierce, Florida t~. Since that time, attemptshave been made to reduce high cost and high energy inputs to the culturesystem, such as continuously pumped seawater nutrients and air, while stillachieving high yields. This approach has required a program of basic re­search on the physiology and nutrition of Gracilaria. and some other speciesof seaweed, in order to determine exactly what the requirements are foroptimal growth of the alga, how those requirements are met by the emperically­derived methods used to obtain the high yields, and how those requirementsmight be met by other, less costly procedures.

Nutrient Uptake Kinetics

~utrient-starved Gracilaria, placed in an enriched seawater solution ineither the light or dark, are able to soak up sufficient ammonium-nitrogenin 6-8 hours to double its internal nitrogen content (i.e., from ca. 1.5to 3.0% of dry weight) and permit its subsequent growth at non-nutrient­limiting rates for one to two weeks (depending upon the growth rate as

III

determined by light, temperature and other factors) in unenriched, flowingseawater. N03-N is taken up less efficiently than is NH4-N (Figure 1) ~J.

!

:\0(

40! ,

20 30

BIWi£KLY £N/UCHMENr /hcuf'S/

t

10

/_---0 _

- -._-0------0-__ .",.-----

I -----0..".--I

II

II

//

I

r1I

II

5 I/

f~ --~---f:.--t::_-_;O-~oe-o

20

F1sun 1. titAN "CttUl OF It"tTllOGtN-S'rARVEIl c;;u.CltAlHA I~ItSE':) III :''ll'Tltrtm ~Olt1f

COh'TAIlt'll:G AlOOlltl.'I'l-NIntOC::N rcl.OSEO CtRC"..!S) 1'1:;0 NtUATt-;:LnOGtll(Ol'tN CIRCUS) tvUY 1',;0 1o""1tS FOR tlit l'to'l.lOllS OF 'tOlE tllDlCA'tC.

Seaweed taken from a flowing seawater culture may thus be removed periodical­ly, soaked for a few hours in a static nutrient solution, and then returnedto the culture system, the process coinciding with the periodic harvestingback of the culture. This would not only conserve nutrients, those unusedremaining in the nutrient bath rather than being lost in the flowing seawatereffluent, but it would also prevent or retard the growth of undesirable epi­phytes brought into the system as spores in the seawater on the cultured sea­weeds and the culture system, a chronic problem in seaweed culture

Initially, the nutrient solution consisted of nitrogen, phosphorus, chelatediron and a mixture of trace metals. Subsequently it was found that the samegrowth was obtained with NH4-N alone as with the complete enrichment medium(Table 1), the unenriched seawater apparently containing an adequate supplyof the other nutrients. Enrichment with nitrogen alone would, of course,represent a considerable cost savings, but whether or not this could be donein other, less nutrient-rich sources of seawater remains to be seen. How­ever, the liquid residues left from anaerobic digestion of Gracilaria formethane production appear to contain all the nutrients necessary for thegrowth of the seaweed and their recycling has been advocated as a low-costnutrient source for biomass production 'D].

112

Table L EFFECT OF EXPOSING mAC!LARIA 24 HOURS PER \·rEEK TO THREE DIFFERENTNUTRIENT ~ffiDIA ON SUBSEQUE~IT GRO\fIH (~ffiAN YIELD: g dry wt/m2. cay) .

Dates Hedia(1981) N N + P N + P + Fe +

trace e1ewents

4/24 - 5/1 12.7 12.9 11. 9

5/1 - 5/8 13.0 14.7 13.0

5/8 - 5/18 21.0 19.3 16.7

5/18 - 5/25 17.0 19.0 16.6

Mean 15.9 16.5 14.6

Effect of Aeration/Mixing

Gracilaria is maintained in suspended culture by aeration along the long axisof the culture tank bottom. What purpose this serves is not known, but appar­ently does not derive from the air itself, since the same growth enhancementis obtained from seaweed kept in suspension by the action of a paddle wheel[4]. It may expose a larger density of seaweed to sunlight than would be pos­sible in an unmixed culture, or it may increase the exposure of the plants toCO 2 and/or other nutrients. Whatever the function, aeration is a major ~ost

and energy input that should be reduced to a minimum level consistent withhigh yield. Table 2 shows that intermittent aeration, for as little as sixhours per day, under two different periodicities, results in the same yieldsof- Gracilaria as does continuous aeration, but that yields decreased in cul­tures aerated for only 5 minutes per hour for a total of two hours per day.The effect of aeration periods intermediate between two and six hours per dayand during daylight hours only will next be investigated.

Table 2. EFFECT OF AERATION ON YIELD OF GRACIL~RIA GROHN AT nroRETENTION TUiES OF ENRICHED FLmnNG SEAl.JATtR O·rEAN YIELDS

4/9-5/4/81 in g dry wt/m 2.day).

Aeration Retention(days)(hours/day) 0.1 1.0

0 18.6 8.5

6 (5 min/hr) 23.0 9.9

6 (15 min 32.9 12.5

6 (2 hrs, 3 x/day, daylight 34.0 14.2

12 (dayligh t) 36.5 14.7

24 37.3 14.7

113

Whatever the effect, it is clear from Table 2 that some degree of aeration orthe mixing it produces is necessary for high sustained yields of the seaweed.Beginning in May, 1981, with the completion of eight new 5 m2 concrete ponds,Gracilaria has been cultured passively on the pond bottom with no mixing oragitation of the seawater, as is done commercially in southern Taiwan CsJ.Yields under different growing conditions (seaweed density, water depth,method of nutrient supply and mixing of the seawater) has resulted to date insome yield differences (Table 3), but in all cases they have been signifi­cantly lower than those obtained in suspended cultures at the same retentiontime (see Table 2)

Table 3. YIELDS FROM POND-BOTTOM CULTURE OF GRACILARIA IN 5 m2 CO~CRETE TANKSWITH SEAWATER RETErITION Tl}lli OF O~~ DAY.

":ater depth Density Nutrient(em) (kg wet wt/m 2) supply

77 3 pulsed

51 3 "26 3 "77 1.5 "77 4.5 "77 6 "77 3 "77 3 continuous

Aeration l

yes

yes

yes

yes

yes

yes

no

yes

Mean yield(g dry wt/m 2.day)

514 - 6/1/81

11.9

9.1

4.5

10.3

8.7

4.9

8.2

8.9

lCentle aeration to mix water but not disturb algae.

Seawater Retention, pH and Carbon Dioxide

The one factor that has been found to date the most important in affectingthe growth and yield of Gracilaria is seawater exchange rate (retention time~

Again the reason is not obvious, but does not appear to be related to thenutrient supply, since the effect is observed when nutrients are added at aconstant rate separately from the seawater supply, or when nutrients aresupplied by soaking the seaweed periodically for a few hours in a static,enriched seawater reservoir, as discussed above. Figure 2 is a summary ofexperiments showing the relationship between yield and retention time, in­cluding cultures ranging in size from 50 to 600,000 liters and with bothcontinuously-enriched and pulse-fed cultures.

From those data and an assumed pumping cost of S2.60/thousand cubic meters($IO/million gallons), it may be estimated that for the methane generatedfrom Gracilaria cultures one meter deep, the cost of seawater pumping alonewould range from $9354/thousand cubic meters ($265/thousand scf) of 100%

114

20,

15!

~CJ

R£T£NT/OIJ (days)

,ecl~-----.l--_----'-----':-------::'o

F:1guz-e 2. YIZlDS OF Ge',}.C!l....?:.t.. AS A ~i.~~c::Ol·! O~ SI:.l"i-iA':!lt R::'I:~"I!C:1 't~.OPEN CnCl.ZS • am::~:UOUS ::~I~'T: ctoszn cracizs • l'TJ!.SED

~","nn::"tS res 2i;. ~iDURS E'lZ~~ 1"'~O ~~.

methane at a retention time of 0.04 days to about $IOO/thousand cubic meters($2.80!thousand scf) of methane at a 20-day retention time (assuming 0.2 1100% methane/g volatile solids). However the a~ea required for the seaweedfarm would range from 40 hectares/l06 m3 of methane produced ~2.8 acres/l06scf) at the 0.04 retention time, to nearly 200 hectares/l06 m methane (14.1acres/lOG scf) at the 20-day retention time. Clearly, it would be desirableto achieve the high yields possible at very rapid exchange rates with muchless water flow. First, however, it is necessary to understand the relation­ship between yield and water exchange

The only essential nutrient not provided in the artificial enrichment nor­mally used to culture Gracilaria and other seaweeds is carbon dioxide. Sea­water of normal salinity (30-35%0 contains about 2 m moles/liter of total C02which exists in the equilibrium:

(1)

Seawater one-meter deep thus contains 24 g C/m2 which could theoreticallysupport the growth of 48 g ash-free dry weight (50% carbon) or about 74 g

115

total dry weight of Gracilaria. a potential yield of 74 g dry wt/m2.day with aretention time of one day. However, removal of free C02 during the photosyn­thetic growth of algae increases the pH; the slower the circulation of sea­water through the seaweed the higher the pH rises (Table 4). At pH > 9.0,there is almost no free CO2 in seawater [6) and its rate of dehydration frombicarbonate to maintain the e9,uilibrium is so slow that it becomes limitingto photosynthesis and growth L~.

Table 4. NEAN, DAYTD!E pH IN GRACIUlRTA CULTUlZES AT DIFFERENT RETEtITIOi'I TUIES.

Time

0800

1030

1315

1435

1530

1615

Retention (days)0.1 1.0

8.2 8.1

8.4 8.6

8.7 9.0

8.7 9.1

8.8 9.1

8.7 9.1

Some seaweeds are able to utilize bicarbonate directly in photosynthesis.Measurement of photosynthesis (~02. by oxygen probe in a closed, recirculat­ing seawater chamber) at four pH levels maintained with TRIS buffer indicate~

however, that Gracilaria can use little or no bicarbonate, with photosynthe~

at pH 9.0 only 19% of that at pH 7.5 (Table 5).

Table 5. EFFECT OF pH ON PHOTOSYNTHESIS OF GRACEMTA Mm ULVA, HE/,StJRED BYOXYGEN INCREASE AFTER 01~ HOlm. FIGURES ARE REL~TIVE TO INCR&\SE AT pH 7.5.

pH Crac i Lar La Uiva

7.5 - 7.6 100 100

8.0 - 8.2 74 1048.6 46 92

9.0 - 9.1 19 92

This, then, would account for the poor growth of Gracilaria at low seawaterexchange rates and at the accompanying high pH levels.

Bicarbonate uptake and its internal dehydration to C02 requires the enzymecarbonic anhydrase. Osterlind [8) has shown that the enzyme is not alwayspresent but must be induced in certain unicellular algae before they canutilize bicarbonate. The question remains of whether carbonic anhydrase

116

can be induced in Gracilaria and/or other species of seaweeds, whether thereare species or clones of Gracilaria that naturally contain the enzyme, orwhether strains of the alga capable of using bacarbonate can be naturallyselected for or genetically produced. In that connection, it would be inter­esting to see whether fast-growing strains of Q. tikvahiae selected by Vander Meer in Halifax [9J may be fast growing because of that ability.

Other species of seaweed do utilize bicarbonate readily and continue photo·synthesis and growth at high pH. The green alga Y1Y! lactuca, was found toproduce oxygen at pH 9.1 at 72% the rate at pH 7.5 (vs. 19% for Gracilaria)(Table 5). Ulva would be an ideal candidate for marine biomass productionfor that reas;n-and because of its high inherent growth rate and potentialyield, were it not for the fact that the alga normally becomes reproductiveand sheds a large fraction of its biomass as microscopic gametes or zoosporesas often as once a week when it is growing rapidly. Dr. Howard Levine (U.Mass.) has kindly provided the author with a sample of Ulva lactuca from apopulation which he has never observed to become reproductive or to bearfruiting bodies. In the several weeks it has since been grown in our cul­ture system, it has also remained sterile, in contrast to several otherstrains of the same species grown under the same conditions. A permanentlysterile clone of Ulv8, as may now be available, could represent an impor­tant contribution to the marine biomass field.

The above observations on bicarbonate utilization by Gracilaria and Ulvaare not original to the present author. Almost exactly the same res~with both species were reported in 1963 by L. R. Blinks [101, who also de­scribed similar experiments with 22 other species of seaweeds. A completespectrum was found in the ability of the different species to utilize bicar­bonate, with y!y! and Gracilaria, among others, representing the two ex­tremes. The giant kelp, Mscrocystis, another popular candidate for marinebiomass production, photosynthesizes only marginally better than Gracilariaat pH 9.0. This basic characteristic of seaweeds may thus be an importantcriterion for the selection of candiadate species for marine biomass pro­duction, if the systems used for such purposes are to consist of extensivemarine farms of high growth rate, high yield algal crops maintained in sea­water that, for logistic or economic reasons, cannot be rapidly exchanged.

REFERENCES

1. Lapointe, B. E. and Ryther, J. H.:of Gracilaria tikvahiae in culture.

Some aspects of the growth and yieldAquaculture, ~ 185-193 (1978).

2. Ryther, J. H., Corwin, N., DeBusk, R. A. and Williams, L. D.: Nitrogenuptake and storage by the red alga Gracilaria tikvahiae. Aquaculture,(in press).

3. Hanisak, M. D.: Recycling the residues from anaerobic digesters as anutrient source for seaweed growth. Bot. Mar., (in press).

117

4. Neish, A. C., Shacklock, P. F., Fox, C. H. and Simpson, F. J.: Thecultivation of Chondrus crispus. Factors affecting growth under green­house conditions. Can. J. Bot., 21, 2263-2271 (1977).

5. Shang, Y. C.: Economic aspects of Graci1aria culture in Taiwan. Aqua­culture, ~ 1-7 (1976).

6. Harvey, H. W.: The chemistry and fertility of sea waters. The Univer­sity Press, Cambridge, 1960, 240 pp.

7. Birmingham, B. C. and Coleman, B.:sation points of freshwater algae.

Measurement of carbon dioxide compen­Plant. Physiol., ~ 892-895 (1979).

8. Osterlind, S.: Inorganic carbon sources of green algae. VI. Furtherexperiments concerning photoactivation of bicarbonate assimilation.Physiol. Plant., 2J 403-408 (1952).

9. Patwary, M. and Van der Meer, J.: Improved biomass yields oftikvahiae through genetic modification of thallus morphology.presented at 20th Northeast Algal Symposium, MBL, Woods Hole,April 11-12, 1981.

Graci1ariaPaper

MA

10. Blinks, L. R.:marine algae.

The effect of pH upon the photosynthesis of littoralPhyco1ogia, ~ 126-136 (1963).

118

WETLAND BIOMASS PRODUCTION

D. C. PrattUniversity of Minnesota Botany Department

1445 Gortner Ave.St. Paul, MN 55108

ABSTRACT

The use of wetlands to produce biomass crops has been the focus of several researchprograms at the University of Minnesota. There are over 6 million hectares of peatlandsin the northern lakes states, Minnesota, Michigan and Wisconsin. Currently only 2.7%of Minnesota peatlands are utilized for crop production. Tvpha )sPp. (cattails), Phrag-mitescommunis (reed grass), Scirpus spp. (rushes), Carex spp. (sedges, and Phalaris arundinacea(reed canarygrass), Alnus spp. (alder) and Salix spp. (willow) are some species beingconsidered as possible wetland energy cropS:-Using wetlands, including peat and wetmineral soils, for the production of energy crops has several advantages:

1) Substantial wetland acreage is available,2) Wetland crops will not be in competition with traditional crops for prime

agricultural lands,3) Wetlands are naturally very productive habitats, often functioning as nutrient

sinks,4) Wetland crops represent a less destructive alternative than peat mining,5) A mixture of native wetland plants can be used, avoiding the traditional

monoeulture approach.

Typha looks particularly promising; it is easily propagated from seed or rhizomes, andis very productive. The total biomass of natural Tvoha stands often exceeds 40 drytons/hectare; yields from stands established from rhizome pieces range from 25-30tons/hectare the first year while stands established from seed yield 8-12 t/ha the firstseason and 15-20 t/ha by the second season.

The current focus for Typha is the development of optimal land preparation, plantingand management techniques. Alternatives are first tested in small replicated fieldplots; the most successful techniques will then be used to establish larger stands.

Studies of harvesting methods, biochemical conversion, land use considerations, possibleenvironmental constraints, and the overall economics of wetland energy crop productionare currently under way, administered by the University's Bio-Energy CoordinatingOffice (BECO). The major long term objective of the cooperative research program isto develop an efficient and renewable energy system using available resources whileminimizing damage to the environment.

BACKGROUND

Productivitv of Wetland PlantsH

1, a..L_

Table 1. STANDING CROP IN NATURAL STANDS OF POTENTIAL WETLAND ENERGYCROPS.

Species Dry Weight, gm-2 Location ReferenceAbove BelowGround Ground Total

CA'ITAILS

Typha ! glauea 1440 2650 4090 Minnesota 62320 2400 4720 Minnesota 41361 New York 72106 Iowa 2

T. latifolia 428-2252 Southeast, U.S. 81400 503 1903 Wisconsin 3500-2000 200-1400 800-3400 Czechoslovakia 9

1::. elephantina 975-2464 1542-5269 India 10t, angustifolia 564-1647 306-2861 4508 USSR 11

1118 England 12

BULRUSH

SCirpus fluviatilis 852 429 1281 Wisconsin 13450 429 1833 Iowa 14466 Iowa 2

REED

Phragmltescommunis 1110 Iowa 14

1118 Minnesota 151115 Czechoslovakia 9

REED CANARYGRASS

Phalarisarundinaeea 1370 New York 16

870 England 171353 Wisconsin 13

SEDGE

Carex atherodes 1160 Minnesota 18

C. lacustris 738 Minnesota 18940 134 1074 Wisconsin 13857 161 1018 New York 191145 New York 20

120

Natural Stands

Wetlands dominated by Typha and other emergent macrophytes are one of the mostproductive natural systems in the temperate zone (1). Above ground standing crops innatural stands often exceed 15 dry tons/hectare (2, 3) while total biomass in the mostproductive stands of over 40 dry tons/hectare have been reported (4, 5). Table 1summarizes productivity estmates of potential wetland crop including cattails, bulrushes,reeds, reed canary grass, and sedges.

Typha is Particularly Attractive as a Biomass Crop. Table 2 presents a summary ofTvpha yields from different regions. Reports of above ground dry weight range from428 g/m 2 to 2464 g/m 2. These data should be interpreted cautiously, however, becausesamples were collected at different times during the season, in different years, usingvarious sampling techniques. In some cases sample size was not sufficiently large toinsure that results are representative of yields over an entire region.

Table 2. CATTAIL YIELDS IN MANAGED STANDS

Dry Weight, gm-2Above Below

Soecies Ground Ground Total Description

T. ! glauea 1486 5455 6941 Rhizomes, l.5m 2 paddiessoil mix + fertilizer,Minnesota (5)

T. ! glauea 810-1540 2670-3200 3670-4210 Rhizomes, 1.5m2 paddies,soil mix + fertilizer, :\lIN (21)

T. ! glauca 1298-1469 1346-1762 2644-3231 Rhizomes, 1.5m2 paddies,peat + fertilizer, MN (22)

T. latifolia 2673-3349 1973-2581 5254-5802 Rhizomes, outdoor hydro-ponte cultures, Czechoslovakia(23)

T. latifolia 920-1200 Transolanted seedling'S,second year, Florida (24)

T. x zlaucn 1268-1448 519-840 1697-2268 Seed, 105m2 paddies, secondyear, MN (22)

121

Typha Yields in Managed Stands

Table 2 presents a summary from the literature of~ productivity in managedstands established with rhizomes, seedlings and seed. Generally total biomass equalsor exceeds figures from natural stands. Obviously factors such as method of standestablishment (rhizomes, seedlings, seed), nutrient availability, planting density andmanagement techniques have a tremendous effect on final yields and the yields obtainedfrom small, carefully managed stands cannot necessarily be achieved in field trials. Afield study was initiated in 1980 to examine some of these variables.

Peatland Stand Establishment Study. Tvoha was established in a northern Minnesotapeatland using rhizomes and seed. The influence of planting density, substratepreparation, and fertilizer application on the seasonal development and productivity oftrial plots was investigated (25). Table 3 summarizes some of the results of this study.Substrate preparation (rotovated or excavated), fertilizer application and increasedplanting density resulted in significantly increased yields. A maximum yield of 16tons/hectare was obtained with a late May planting date using a planting rate of 25rhizomes/m2 with 90, 158 and 300 kg/ha of N, P, and K respectively. Although finalyield increased with planting density stand establishment using relatively low plantingrates may be cost effective.PROJECT OBJECTIVES

A multidisciplinary team of researchers from the Departments of Botany, Ecology andHorticulture at the University of Minnesota is working on wetland crop production.The major objectives are to develop land preparation, propagation, planting andmanagement practices that optimize sustained yield while minimizing energy costs andenvironmental degradation. While the initial focus is on the continued development ofcattails as a commercial crop, several other potential wetland biomass crops will beconsidered including reeds, bulrushes, and sedges.

OUTLINE OF RESEARCH PROGRAM

I. Selection and Propagation of Wetland Biomass CropsA. Identification of Promising SpeciesB. Selection of Highly Productive IndividualsC. Micropropagation StudiesD. Maeropropagatton StudiesE. Production of Commercial Scale Quantities

II. Field TrialsA. Site Preparation MethodsB. Planting TechniquesC. Fertilizer RequirementsD. Insect and Weed Control TechniquesE. Harvesting Schedule and Regrowth

III. Establishment of Large Plots (Year 2)A. Site SelectionB. Site Preparation and PlantingC. Monitor GrowthD. Monitor Environmental Effects

TABLE 3 EFFECTS OF TREATMENTS ON STANDING CROP AND DENSITya

Effect of Planting DensityArea tested

T. latifoliaRotovated ,fertilized

T. ! glaucaRotovated ,fertilizedT. latifoliaExcavated ,fertilized

Density vs densi tv

91m2 vs 251m241m2 vs 251m211m2 vs 251m 241m 2 vs 91m 211m2 vs 91m 211m2 vs 41m2

11m2 vs 4/m 2b

41m2 vs 91m211m2 vs 91m 211m2 vs 41m2

Standing crop

sS

SS

sns

ns

nsss

Final densitvy

S

55

nssns

ns

nssns

Effect of Substrate PreparationRotovated vs excavated

T. latifolia, fertilized1':" latifolia, fertilizedT. latifolia, fertilized

Rotovated vs unrotovatedt:. latifolia, fertilized

T. latifolia, unfertilizedT. ! glauca, fertilized

Effect of FertilizerFertilized vs nonfertilized

T. latifoliaT. latifolia

Effect of Species Type!:.latifolia vs T. ! glauca

Rotova ted, fertilizedRotovated, fertilized

91m2

4/m2b1/m2b

91m2

91m21/m2b

91m291m 2

4/m2b1/m 2b

nsnsns

s

ss

S

5

nsns

nsosns

ns

nss

nsns

nsns

as = significant difference, ns = no significant difference, = 0.05.bT-test.

123

CURRENT FOCUS

Selection

Identifiea tion of Promising Genotypes

Cattail rhizomes were collected from productive natural stands identified from previousstudies (26). The rhizomes were planted in peat-filled paddies using a 6 x 6 latinsquare design in order to compare productivity under identical growing conditions. Thisexperiment will also provide us with sufficient planting material for more extensiveexperiments in subsequent seasons.

Propagation

The three stages in the propagation work include:1. Establishment of Typha explants in tissue culture with formation of callus tissue.2. Proliferation of individual plantlets in tissue culture, and3. Establishment of these plantlets in soil and then in the field.Current work with Typha is now at step 1.

Attempts to micropropagate~ began in December, 1980. Four trials have beenconducted, averaging 160 explants each. Stock plants have also been grown from seedand rhizomes, some under sterile conditions, to serve as explant sources. There are alarge number of factors which must be considered in starting micropropagation workwith a new species. Some of these factors and some of the possible alternatives whichare being used with Typha include: 1) Part of plant being used as the explant, 2) Culturemedium, 3) Plant growth regulators, 4. Environmental conditions, 5) State of medium,and 6) Decontamination procedures.

Establishment of Typha rhizome explants in a sterile tissue culture environment hasbeen slow. Explants survive longest when kept in the dark. The parts of stock plantswhich have successfully grown in culture are buds, germinating seeds, and intercalarymeristems of young leaves. The medium/growth regulator combination which hasproduced the best growth so far is Linsmaiser and Skoog medium with 5 rng/l 2,4-D.Less contamination occurs when explants are pre-soaked in sucrose and then a fungicide.

Field Trials

Land preparation, planting and management alternatives are first being tested in smallreplicated field plots. Variables being examined include type of planting material(rhizomes, seedlings, seed), fertilizer application rates, timing of fertilizer application,machine planting and depth of peat removal The most successful techniques will beused to establish larger stands.The objectives of the first three experiments are todetermine the effect of fertilizer applications on Tyoha yields in peat soils, the rateof nutrient uptake over the growing season and the effects of timing the applicationof fertilizer to coincide with maximum nutrient uptake. Since space limitations precludethe presentation of a detailed description of each experiment, Experiment G8101 willserve as an example.

Experiment G8101 - Fertilization Study

Objectives:

124

1) Determine the effect of different combinations of fertilizer on the amount of aboveand below ground biomass produced after one and two years on a previously uncultivatedpeat soil.2) Determine the nutrient standing crop after one and two years.3) Determine the effect of different combinations of fertilizer on density increase andshoot height over the course of two growing seasons.4) Determine the effect of different combinations of fertilizer on soil fertility afterone and two years.5) Provide a source of above ground biomass for experiments involving harvesting andhandling of biomass.

Design:1) Blocked complete factorial (3 x 2 x 2)2) Factors: Nitrogen (3 levels), Phosphorus (2 levels), Potassium (2 levels)3) Levels: a) Nitrogen - 0, 75, and 150 kg (elemental form)

Phosphorus - 0, 150 kg/ha (oxide form)Potassium - 0, 300 kg!ha (oxide form)

4) Treatments per block: 125) Blocks: 46) Total Plots: 487) Size of Plots: 3 m x 5 m8) Planting Density: 9 rhizomes/m 29) Field Prerparation: a) field wa.s disked 3" deep and dragged, b) 70 kg/ha of PetersFritted Trace Elements, c) 20 kg/he. of CUS04, d) plots were fertilized and thenrotovated to a depth of 6".

Observations and Samoling::1) Density2) Height3) Insect Damage4) Competitors5) Water level6) Soil samples for nutrient and bulk density analysis7) Above and below ground plant samples for biomass determination, and tissue nutrientanalysis.

REFERENCES

(1) Westlake, D.F.: Comparisons of Plant Productivity. BioI Rev., 38, 385-425 (1963).(2) van der Vall<, A.G. and Davis, C.B.: Primary Production of Prairie Glacial Marshes.

In: Good, R.E. et ale (eds.), Freshwater Wetlands: Production Processes andManagement Potential, Academic Press, NY, 1978.

(3) Gustafson, T.D.: Production, Photosynthesis and the Storage and Utilization ofReserves in a Natural Stand of Tvoha latifolia L. Ph.D. Thesis, University ofWisconsin, Madison, 102 pp (1976"""---

(4) Andrews, N.J and Pratt, D.C.: The Potential of Cattails (Tvoha spp.) as an EnergySource: Productivity in Managed Stands. J. }jlinn. Acad, Sci., 44, 5-8 (1978).

(5) Fox, C.A.: Capture of Radiant Energy by Plants. M.S. Thesis, University of Minnesota,49 pp (1975).

(6) Bray, J.R.: The Chlorophyll Content of Some Native and Managed Plant Communitiesin Central Minnesota. Can. J. Bot., 38, 313-333 (1960).

(7) Bernard, J.M. and Fitz, M.L.: Seasonal Changes in Above-ground Primary Productionand Nutrient Contents in a Central New York Tvoha glauca Ecosystem. Bull.

125

Torrey Bot. Club, 106, 37-40 (1978).(8) Boyd, C.E. and Hess, L.W.: Factors Influencing Shoot Production and Mineral Nutrient

Levels in TYpha latifolia. Ecology 51, 296-300 (1970).(9) Kvet, J. and Husak, S.: Primary Data on Biomass Production Estimates in Typical

Stands of Fishpond Littoral Plant Communities. In: Dykyjova, D. and Kvet, J.(eds.) Pond Littoral Ecosystems, Springer-Verlag, NY (1978).

(10) Sharma, K.P. and B. Gopal: Studies on Stand Structure and Primary Production inTypha species. Int. J. Beol. Environ. Sei., 45-66 (1977).

(11) Merezhko, A.I., N.H. Smimova and V.P. Gorbik: Growth of Stands of LesserReedmace (~an~stifolia)and the Functioning of its Root System. Gidrobiol.ZH., 15, 20=25'{I979).

(12) Mason, CP. and R.J. Bryant: Production, Nutrient Content, and Decomposition ofPhragmites communis Trin. and Typha latifolia L. J. Ecol., 63, 71-96 (1975).

(13) Klopatek, J:M. and Steams, F.W.: Primary Productivity of Emergent Macrophytes ina Wisconsin Freshwater Marsh Ecosystem. Amer. MidI. Nat., 100, 320-333 (1978).

(14) Van Dyke, G.D.: Aspects Relating to Emergent Vegetation Dynamics in a Deep Marsh,. Northcentral Iowa. Ph.D. Thesis, Iowa State University, Ames (1972).

(15) Andrews, N.J.: Unpublished,(16) Wedin, W.F. and Helsel, Z.: Plant Species for Biomass Production on Managed Sites.

Biomass - A Cash Crop for the Future, Kansas City, MO, Proceedings (1977).(17) Pearsall, W.H. and Gorham, E.: Production Ecology. I. Standing Crops of Natural

Vegetation. Oikos, 7, 193-201 (1956).(18) Gorham, E. and Bernard7 J.M.: Midsummer Standing Crops of Wetland Sedge Meadows

along a Transect from Forest to Prairie. J. Minn. Aead, Sci. 41, 16-17 (1975).(19) Bernard, J.M. and MacDonald, J.G.: Primary Production and Life History of Carex

laeustris. Can. J. acr., 52, 117-123 (1974).(20) Bernard, J.M. and Solsky, B.A.: Nutrient Cycling in a Carex lacustris Wetland.

Can. J. Bot., 55, 630-638 (1970).(21) Moss, D.N.: Improvement of Plant Photosynthesis through Genetic Engineering.

Symposium Papers, Clean Fuels from Biomass and Wastes, I.G.T., 63-71 (1977).(22) Pratt, D.C., Bonnewell, V., Andrews, N.J. and Kim, J.: The Potential of Cattails as

an Energy Source. Final Report to the Minnesota Energy Agency (1980).(23) Dykyjova, D., Veber, K. and Priban, K.: Productivity and Root/Shoot Ratio of

Reedswamp Species Growing in Outdoor Hydroponic Cultures. Folio Geobot.Phytotax., Praha, 6 (2), 233-254 (1971).

(24) White, J.M. and Sinclair, L.R.: Effect of Plant Spacing on Growth and Yield ofTransplanted Cattails. Proe. Soil Crop Sci. Soc. Fla., 38, 18-20 (1979).

(25) Andrews, N.J., Penko, M., Mattson, M.D., and Pratt, D.C.: The Establishment ofCattails on a Minnesota Peatland. Report to the Minn. Department of NaturalResources, 100 pp. .

(26) Bonnewell, V.: Typha Productivity, Mineral Nutrition, and Seed Germination. Ph.D.Thesis, University of Minnesota, 132 pp (1981).

126